Merge pull request #35 from justadudewhohacks/batch-input

build/NetInput.d.ts

-import { Dimensions, TMediaElement } from './types';
+import * as tf from '@tensorflow/tfjs-core';
+import { Point } from './Point';
+import { TResolvedNetInput } from './types';
 export declare class NetInput {
-    private _canvases;
-    constructor(medias: Array<TMediaElement>, dims?: Dimensions);
-    private initCanvas(media, dims?);
-    readonly canvases: HTMLCanvasElement[];
-    readonly width: number;
-    readonly height: number;
-    readonly dims: Dimensions | null;
+    private _inputs;
+    private _isManaged;
+    private _isBatchInput;
+    private _inputDimensions;
+    private _paddings;
+    constructor(inputs: tf.Tensor4D | Array<TResolvedNetInput>, isBatchInput?: boolean);
+    readonly inputs: tf.Tensor3D[];
+    readonly isManaged: boolean;
+    readonly isBatchInput: boolean;
+    readonly batchSize: number;
+    readonly inputDimensions: number[][];
+    readonly paddings: Point[];
+    getInputDimensions(batchIdx: number): number[];
+    getInputHeight(batchIdx: number): number;
+    getInputWidth(batchIdx: number): number;
+    getPaddings(batchIdx: number): Point;
+    toBatchTensor(inputSize: number, isCenterInputs?: boolean): tf.Tensor4D;
+    /**
+     *  By setting the isManaged flag, all newly created tensors will be automatically
+     *  automatically disposed after the batch tensor has been created
+     */
+    managed(): this;
+    dispose(): void;
 }

build/NetInput.js

 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
+var tf = require("@tensorflow/tfjs-core");
+var isTensor_1 = require("./commons/isTensor");
+var padToSquare_1 = require("./padToSquare");
+var Point_1 = require("./Point");
 var utils_1 = require("./utils");
 var NetInput = /** @class */ (function () {
-    function NetInput(medias, dims) {
-        var _this = this;
-        this._canvases = [];
-        medias.forEach(function (m) { return _this.initCanvas(m, dims); });
-    }
-    NetInput.prototype.initCanvas = function (media, dims) {
-        if (media instanceof HTMLCanvasElement) {
-            this._canvases.push(media);
-            return;
+    function NetInput(inputs, isBatchInput) {
+        if (isBatchInput === void 0) { isBatchInput = false; }
+        this._inputs = [];
+        this._isManaged = false;
+        this._isBatchInput = false;
+        this._inputDimensions = [];
+        this._paddings = [];
+        if (isTensor_1.isTensor4D(inputs)) {
+            this._inputs = tf.unstack(inputs);
         }
-        // if input is batch type, make sure every canvas has the same dimensions
-        var canvasDims = this.dims || dims;
-        this._canvases.push(utils_1.createCanvasFromMedia(media, canvasDims));
-    };
-    Object.defineProperty(NetInput.prototype, "canvases", {
+        if (Array.isArray(inputs)) {
+            this._inputs = inputs.map(function (input) {
+                if (isTensor_1.isTensor3D(input)) {
+                    // TODO: make sure not to dispose original tensors passed in by the user
+                    return tf.clone(input);
+                }
+                return tf.fromPixels(input instanceof HTMLCanvasElement ? input : utils_1.createCanvasFromMedia(input));
+            });
+        }
+        this._isBatchInput = this.batchSize > 1 || isBatchInput;
+        this._inputDimensions = this._inputs.map(function (t) { return t.shape; });
+    }
+    Object.defineProperty(NetInput.prototype, "inputs", {
+        get: function () {
+            return this._inputs;
+        },
+        enumerable: true,
+        configurable: true
+    });
+    Object.defineProperty(NetInput.prototype, "isManaged", {
         get: function () {
-            return this._canvases;
+            return this._isManaged;
         },
         enumerable: true,
         configurable: true
     });
-    Object.defineProperty(NetInput.prototype, "width", {
+    Object.defineProperty(NetInput.prototype, "isBatchInput", {
         get: function () {
-            return (this._canvases[0] || {}).width;
+            return this._isBatchInput;
         },
         enumerable: true,
         configurable: true
     });
-    Object.defineProperty(NetInput.prototype, "height", {
+    Object.defineProperty(NetInput.prototype, "batchSize", {
         get: function () {
-            return (this._canvases[0] || {}).height;
+            return this._inputs.length;
         },
         enumerable: true,
         configurable: true
     });
-    Object.defineProperty(NetInput.prototype, "dims", {
+    Object.defineProperty(NetInput.prototype, "inputDimensions", {
         get: function () {
-            var _a = this, width = _a.width, height = _a.height;
-            return (width > 0 && height > 0) ? { width: width, height: height } : null;
+            return this._inputDimensions;
         },
         enumerable: true,
         configurable: true
     });
+    Object.defineProperty(NetInput.prototype, "paddings", {
+        get: function () {
+            return this._paddings;
+        },
+        enumerable: true,
+        configurable: true
+    });
+    NetInput.prototype.getInputDimensions = function (batchIdx) {
+        return this._inputDimensions[batchIdx];
+    };
+    NetInput.prototype.getInputHeight = function (batchIdx) {
+        return this._inputDimensions[batchIdx][0];
+    };
+    NetInput.prototype.getInputWidth = function (batchIdx) {
+        return this._inputDimensions[batchIdx][1];
+    };
+    NetInput.prototype.getPaddings = function (batchIdx) {
+        return this._paddings[batchIdx];
+    };
+    NetInput.prototype.toBatchTensor = function (inputSize, isCenterInputs) {
+        var _this = this;
+        if (isCenterInputs === void 0) { isCenterInputs = true; }
+        return tf.tidy(function () {
+            var inputTensors = _this._inputs.map(function (inputTensor) {
+                var _a = inputTensor.shape, originalHeight = _a[0], originalWidth = _a[1];
+                var imgTensor = inputTensor.expandDims().toFloat();
+                imgTensor = padToSquare_1.padToSquare(imgTensor, isCenterInputs);
+                var _b = imgTensor.shape.slice(1), heightAfterPadding = _b[0], widthAfterPadding = _b[1];
+                if (heightAfterPadding !== inputSize || widthAfterPadding !== inputSize) {
+                    imgTensor = tf.image.resizeBilinear(imgTensor, [inputSize, inputSize]);
+                }
+                _this._paddings.push(new Point_1.Point(widthAfterPadding - originalWidth, heightAfterPadding - originalHeight));
+                return imgTensor;
+            });
+            var batchTensor = tf.stack(inputTensors).as4D(_this.batchSize, inputSize, inputSize, 3);
+            if (_this.isManaged) {
+                _this.dispose();
+            }
+            return batchTensor;
+        });
+    };
+    /**
+     *  By setting the isManaged flag, all newly created tensors will be automatically
+     *  automatically disposed after the batch tensor has been created
+     */
+    NetInput.prototype.managed = function () {
+        this._isManaged = true;
+        return this;
+    };
+    NetInput.prototype.dispose = function () {
+        this._inputs.forEach(function (t) { return t.dispose(); });
+    };
     return NetInput;
 }());
 exports.NetInput = NetInput;

build/NetInput.js.map

-{"version":3,"file":"NetInput.js","sourceRoot":"","sources":["../src/NetInput.ts"],"names":[],"mappings":";;AACA,iCAAgD;AAEhD;IAGE,kBACE,MAA4B,EAC5B,IAAiB;QAFnB,iBAMC;QAFC,IAAI,CAAC,SAAS,GAAG,EAAE,CAAA;QACnB,MAAM,CAAC,OAAO,CAAC,UAAA,CAAC,IAAI,OAAA,KAAI,CAAC,UAAU,CAAC,CAAC,EAAE,IAAI,CAAC,EAAxB,CAAwB,CAAC,CAAA;IAC/C,CAAC;IAEO,6BAAU,GAAlB,UAAmB,KAAoB,EAAE,IAAiB;QACxD,IAAI,KAAK,YAAY,iBAAiB,EAAE;YACtC,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YAC1B,OAAM;SACP;QAED,yEAAyE;QACzE,IAAM,UAAU,GAAG,IAAI,CAAC,IAAI,IAAI,IAAI,CAAA;QACpC,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC,6BAAqB,CAAC,KAAK,EAAE,UAAU,CAAC,CAAC,CAAA;IAC/D,CAAC;IAED,sBAAW,8BAAQ;aAAnB;YACE,OAAO,IAAI,CAAC,SAAS,CAAA;QACvB,CAAC;;;OAAA;IAED,sBAAW,2BAAK;aAAhB;YACE,OAAO,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,KAAK,CAAA;QACxC,CAAC;;;OAAA;IAED,sBAAW,4BAAM;aAAjB;YACE,OAAO,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,MAAM,CAAA;QACzC,CAAC;;;OAAA;IAED,sBAAW,0BAAI;aAAf;YACQ,IAAA,SAAwB,EAAtB,gBAAK,EAAE,kBAAM,CAAS;YAC9B,OAAO,CAAC,KAAK,GAAG,CAAC,IAAI,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,KAAK,OAAA,EAAE,MAAM,QAAA,EAAE,CAAC,CAAC,CAAC,IAAI,CAAA;QAC7D,CAAC;;;OAAA;IACH,eAAC;AAAD,CAAC,AAtCD,IAsCC;AAtCY,4BAAQ"}
\ No newline at end of file
+{"version":3,"file":"NetInput.js","sourceRoot":"","sources":["../src/NetInput.ts"],"names":[],"mappings":";;AAAA,0CAA4C;AAE5C,+CAA4D;AAC5D,6CAA4C;AAC5C,iCAAgC;AAEhC,iCAAgD;AAEhD;IAQE,kBACE,MAA8C,EAC9C,YAA6B;QAA7B,6BAAA,EAAA,oBAA6B;QATvB,YAAO,GAAkB,EAAE,CAAA;QAC3B,eAAU,GAAY,KAAK,CAAA;QAC3B,kBAAa,GAAY,KAAK,CAAA;QAE9B,qBAAgB,GAAe,EAAE,CAAA;QACjC,cAAS,GAAY,EAAE,CAAA;QAM7B,IAAI,qBAAU,CAAC,MAAM,CAAC,EAAE;YACtB,IAAI,CAAC,OAAO,GAAG,EAAE,CAAC,OAAO,CAAC,MAAqB,CAAkB,CAAA;SAClE;QAED,IAAI,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC,EAAE;YACzB,IAAI,CAAC,OAAO,GAAG,MAAM,CAAC,GAAG,CAAC,UAAA,KAAK;gBAC7B,IAAI,qBAAU,CAAC,KAAK,CAAC,EAAE;oBACrB,wEAAwE;oBACxE,OAAO,EAAE,CAAC,KAAK,CAAC,KAAoB,CAAC,CAAA;iBACtC;gBAED,OAAO,EAAE,CAAC,UAAU,CAClB,KAAK,YAAY,iBAAiB,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,6BAAqB,CAAC,KAA4C,CAAC,CACjH,CAAA;YACH,CAAC,CAAC,CAAA;SACH;QAED,IAAI,CAAC,aAAa,GAAG,IAAI,CAAC,SAAS,GAAG,CAAC,IAAI,YAAY,CAAA;QACvD,IAAI,CAAC,gBAAgB,GAAG,IAAI,CAAC,OAAO,CAAC,GAAG,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,CAAC,KAAK,EAAP,CAAO,CAAC,CAAA;IACxD,CAAC;IAED,sBAAW,4BAAM;aAAjB;YACE,OAAO,IAAI,CAAC,OAAO,CAAA;QACrB,CAAC;;;OAAA;IAED,sBAAW,+BAAS;aAApB;YACE,OAAO,IAAI,CAAC,UAAU,CAAA;QACxB,CAAC;;;OAAA;IAED,sBAAW,kCAAY;aAAvB;YACE,OAAO,IAAI,CAAC,aAAa,CAAA;QAC3B,CAAC;;;OAAA;IAED,sBAAW,+BAAS;aAApB;YACE,OAAO,IAAI,CAAC,OAAO,CAAC,MAAM,CAAA;QAC5B,CAAC;;;OAAA;IAED,sBAAW,qCAAe;aAA1B;YACE,OAAO,IAAI,CAAC,gBAAgB,CAAA;QAC9B,CAAC;;;OAAA;IAED,sBAAW,8BAAQ;aAAnB;YACE,OAAO,IAAI,CAAC,SAAS,CAAA;QACvB,CAAC;;;OAAA;IAEM,qCAAkB,GAAzB,UAA0B,QAAgB;QACxC,OAAO,IAAI,CAAC,gBAAgB,CAAC,QAAQ,CAAC,CAAA;IACxC,CAAC;IAEM,iCAAc,GAArB,UAAsB,QAAgB;QACpC,OAAO,IAAI,CAAC,gBAAgB,CAAC,QAAQ,CAAC,CAAC,CAAC,CAAC,CAAA;IAC3C,CAAC;IAEM,gCAAa,GAApB,UAAqB,QAAgB;QACnC,OAAO,IAAI,CAAC,gBAAgB,CAAC,QAAQ,CAAC,CAAC,CAAC,CAAC,CAAA;IAC3C,CAAC;IAEM,8BAAW,GAAlB,UAAmB,QAAgB;QACjC,OAAO,IAAI,CAAC,SAAS,CAAC,QAAQ,CAAC,CAAA;IACjC,CAAC;IAEM,gCAAa,GAApB,UAAqB,SAAiB,EAAE,cAA8B;QAAtE,iBA+BC;QA/BuC,+BAAA,EAAA,qBAA8B;QAEpE,OAAO,EAAE,CAAC,IAAI,CAAC;YAEb,IAAM,YAAY,GAAG,KAAI,CAAC,OAAO,CAAC,GAAG,CAAC,UAAC,WAAwB;gBACvD,IAAA,sBAAmD,EAAlD,sBAAc,EAAE,qBAAa,CAAqB;gBAEzD,IAAI,SAAS,GAAG,WAAW,CAAC,UAAU,EAAE,CAAC,OAAO,EAAiB,CAAA;gBACjE,SAAS,GAAG,yBAAW,CAAC,SAAS,EAAE,cAAc,CAAC,CAAA;gBAE5C,IAAA,6BAAkE,EAAjE,0BAAkB,EAAE,yBAAiB,CAA4B;gBAExE,IAAI,kBAAkB,KAAK,SAAS,IAAI,iBAAiB,KAAK,SAAS,EAAE;oBACvE,SAAS,GAAG,EAAE,CAAC,KAAK,CAAC,cAAc,CAAC,SAAS,EAAE,CAAC,SAAS,EAAE,SAAS,CAAC,CAAC,CAAA;iBACvE;gBAED,KAAI,CAAC,SAAS,CAAC,IAAI,CAAC,IAAI,aAAK,CAC3B,iBAAiB,GAAG,aAAa,EACjC,kBAAkB,GAAG,cAAc,CACpC,CAAC,CAAA;gBACF,OAAO,SAAS,CAAA;YAClB,CAAC,CAAC,CAAA;YAEF,IAAM,WAAW,GAAG,EAAE,CAAC,KAAK,CAAC,YAAY,CAAC,CAAC,IAAI,CAAC,KAAI,CAAC,SAAS,EAAE,SAAS,EAAE,SAAS,EAAE,CAAC,CAAC,CAAA;YAExF,IAAI,KAAI,CAAC,SAAS,EAAE;gBAClB,KAAI,CAAC,OAAO,EAAE,CAAA;aACf;YAED,OAAO,WAAW,CAAA;QACpB,CAAC,CAAC,CAAA;IACJ,CAAC;IAED;;;OAGG;IACI,0BAAO,GAAd;QACE,IAAI,CAAC,UAAU,GAAG,IAAI,CAAA;QACtB,OAAO,IAAI,CAAA;IACb,CAAC;IAEM,0BAAO,GAAd;QACE,IAAI,CAAC,OAAO,CAAC,OAAO,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,CAAC,OAAO,EAAE,EAAX,CAAW,CAAC,CAAA;IACxC,CAAC;IACH,eAAC;AAAD,CAAC,AAtHD,IAsHC;AAtHY,4BAAQ"}
\ No newline at end of file

build/allFacesFactory.d.ts

-import * as tf from '@tensorflow/tfjs-core';
 import { FaceDetectionNet } from './faceDetectionNet/FaceDetectionNet';
 import { FaceLandmarkNet } from './faceLandmarkNet/FaceLandmarkNet';
 import { FaceRecognitionNet } from './faceRecognitionNet/FaceRecognitionNet';
 import { FullFaceDescription } from './FullFaceDescription';
-import { NetInput } from './NetInput';
-export declare function allFacesFactory(detectionNet: FaceDetectionNet, landmarkNet: FaceLandmarkNet, recognitionNet: FaceRecognitionNet): (input: string | HTMLCanvasElement | HTMLImageElement | HTMLVideoElement | (string | HTMLCanvasElement | HTMLImageElement | HTMLVideoElement)[] | tf.Tensor<tf.Rank> | NetInput, minConfidence: number) => Promise<FullFaceDescription[]>;
+import { TNetInput } from './types';
+export declare function allFacesFactory(detectionNet: FaceDetectionNet, landmarkNet: FaceLandmarkNet, recognitionNet: FaceRecognitionNet): (input: TNetInput, minConfidence: number) => Promise<FullFaceDescription[]>;

build/allFacesFactory.js

@@ -12,21 +12,25 @@ function allFacesFactory(detectionNet, landmarkNet, recognitionNet) {
                     case 0: return [4 /*yield*/, detectionNet.locateFaces(input, minConfidence)];
                     case 1:
                         detections = _a.sent();
-                        return [4 /*yield*/, extractFaceTensors_1.extractFaceTensors(input, detections)];
+                        return [4 /*yield*/, extractFaceTensors_1.extractFaceTensors(input, detections)
+                            /**
+                            const faceLandmarksByFace = await Promise.all(faceTensors.map(
+                              faceTensor => landmarkNet.detectLandmarks(faceTensor)
+                            )) as FaceLandmarks[]
+                             */
+                        ];
                     case 2:
                         faceTensors = _a.sent();
-                        return [4 /*yield*/, Promise.all(faceTensors.map(function (faceTensor) { return landmarkNet.detectLandmarks(faceTensor); }))];
+                        return [4 /*yield*/, landmarkNet.detectLandmarks(faceTensors)];
                     case 3:
                         faceLandmarksByFace = _a.sent();
                         faceTensors.forEach(function (t) { return t.dispose(); });
-                        return [4 /*yield*/, Promise.all(faceLandmarksByFace.map(function (landmarks, i) { return landmarks.align(detections[i].getBox()); }))];
-                    case 4:
-                        alignedFaceBoxes = _a.sent();
+                        alignedFaceBoxes = faceLandmarksByFace.map(function (landmarks, i) { return landmarks.align(detections[i].getBox()); });
                         return [4 /*yield*/, extractFaceTensors_1.extractFaceTensors(input, alignedFaceBoxes)];
-                    case 5:
+                    case 4:
                         alignedFaceTensors = _a.sent();
                         return [4 /*yield*/, Promise.all(alignedFaceTensors.map(function (faceTensor) { return recognitionNet.computeFaceDescriptor(faceTensor); }))];
-                    case 6:
+                    case 5:
                         descriptors = _a.sent();
                         alignedFaceTensors.forEach(function (t) { return t.dispose(); });
                         return [2 /*return*/, detections.map(function (detection, i) {

build/allFacesFactory.js.map

-{"version":3,"file":"allFacesFactory.js","sourceRoot":"","sources":["../src/allFacesFactory.ts"],"names":[],"mappings":";;;AAEA,2DAA0D;AAI1D,6DAA4D;AAI5D,yBACE,YAA8B,EAC9B,WAA4B,EAC5B,cAAkC;IAElC,OAAO,UACL,KAAuC,EACvC,aAAqB;;;;;4BAGF,qBAAM,YAAY,CAAC,WAAW,CAAC,KAAK,EAAE,aAAa,CAAC,EAAA;;wBAAjE,UAAU,GAAG,SAAoD;wBAEnD,qBAAM,uCAAkB,CAAC,KAAK,EAAE,UAAU,CAAC,EAAA;;wBAAzD,WAAW,GAAG,SAA2C;wBACnC,qBAAM,OAAO,CAAC,GAAG,CAAC,WAAW,CAAC,GAAG,CAC3D,UAAA,UAAU,IAAI,OAAA,WAAW,CAAC,eAAe,CAAC,UAAU,CAAC,EAAvC,CAAuC,CACtD,CAAC,EAAA;;wBAFI,mBAAmB,GAAG,SAE1B;wBACF,WAAW,CAAC,OAAO,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,CAAC,OAAO,EAAE,EAAX,CAAW,CAAC,CAAA;wBAEZ,qBAAM,OAAO,CAAC,GAAG,CAAC,mBAAmB,CAAC,GAAG,CAChE,UAAC,SAAS,EAAE,CAAC,IAAK,OAAA,SAAS,CAAC,KAAK,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAAC,EAAvC,CAAuC,CAC1D,CAAC,EAAA;;wBAFI,gBAAgB,GAAG,SAEvB;wBACyB,qBAAM,uCAAkB,CAAC,KAAK,EAAE,gBAAgB,CAAC,EAAA;;wBAAtE,kBAAkB,GAAG,SAAiD;wBAExD,qBAAM,OAAO,CAAC,GAAG,CAAC,kBAAkB,CAAC,GAAG,CAC1D,UAAA,UAAU,IAAI,OAAA,cAAc,CAAC,qBAAqB,CAAC,UAAU,CAAC,EAAhD,CAAgD,CAC/D,CAAC,EAAA;;wBAFI,WAAW,GAAG,SAElB;wBACF,kBAAkB,CAAC,OAAO,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,CAAC,OAAO,EAAE,EAAX,CAAW,CAAC,CAAA;wBAE5C,sBAAO,UAAU,CAAC,GAAG,CAAC,UAAC,SAAS,EAAE,CAAC;gCACjC,OAAA,IAAI,yCAAmB,CACrB,SAAS,EACT,mBAAmB,CAAC,CAAC,CAAC,CAAC,YAAY,CAAC,SAAS,CAAC,MAAM,EAAE,CAAC,EACvD,WAAW,CAAC,CAAC,CAAC,CACf;4BAJD,CAIC,CACF,EAAA;;;;KAEF,CAAA;AACH,CAAC;AArCD,0CAqCC"}
\ No newline at end of file
+{"version":3,"file":"allFacesFactory.js","sourceRoot":"","sources":["../src/allFacesFactory.ts"],"names":[],"mappings":";;;AAAA,2DAA0D;AAK1D,6DAA4D;AAG5D,yBACE,YAA8B,EAC9B,WAA4B,EAC5B,cAAkC;IAElC,OAAO,UACL,KAAgB,EAChB,aAAqB;;;;;4BAGF,qBAAM,YAAY,CAAC,WAAW,CAAC,KAAK,EAAE,aAAa,CAAC,EAAA;;wBAAjE,UAAU,GAAG,SAAoD;wBAEnD,qBAAM,uCAAkB,CAAC,KAAK,EAAE,UAAU,CAAC;4BAC/D;;;;+BAIG;0BAL4D;;wBAAzD,WAAW,GAAG,SAA2C;wBAMnC,qBAAM,WAAW,CAAC,eAAe,CAAC,WAAW,CAAC,EAAA;;wBAApE,mBAAmB,GAAG,SAAiE;wBAE7F,WAAW,CAAC,OAAO,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,CAAC,OAAO,EAAE,EAAX,CAAW,CAAC,CAAA;wBAE/B,gBAAgB,GAAG,mBAAmB,CAAC,GAAG,CAC9C,UAAC,SAAS,EAAE,CAAC,IAAK,OAAA,SAAS,CAAC,KAAK,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAAC,EAAvC,CAAuC,CAC1D,CAAA;wBAC0B,qBAAM,uCAAkB,CAAC,KAAK,EAAE,gBAAgB,CAAC,EAAA;;wBAAtE,kBAAkB,GAAG,SAAiD;wBAExD,qBAAM,OAAO,CAAC,GAAG,CAAC,kBAAkB,CAAC,GAAG,CAC1D,UAAA,UAAU,IAAI,OAAA,cAAc,CAAC,qBAAqB,CAAC,UAAU,CAAC,EAAhD,CAAgD,CAC/D,CAAC,EAAA;;wBAFI,WAAW,GAAG,SAElB;wBACF,kBAAkB,CAAC,OAAO,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,CAAC,OAAO,EAAE,EAAX,CAAW,CAAC,CAAA;wBAE5C,sBAAO,UAAU,CAAC,GAAG,CAAC,UAAC,SAAS,EAAE,CAAC;gCACjC,OAAA,IAAI,yCAAmB,CACrB,SAAS,EACT,mBAAmB,CAAC,CAAC,CAAC,CAAC,YAAY,CAAC,SAAS,CAAC,MAAM,EAAE,CAAC,EACvD,WAAW,CAAC,CAAC,CAAC,CACf;4BAJD,CAIC,CACF,EAAA;;;;KAEF,CAAA;AACH,CAAC;AAzCD,0CAyCC"}
\ No newline at end of file

build/commons/getImageTensor.d.ts

-import * as tf from '@tensorflow/tfjs-core';
-import { NetInput } from '../NetInput';
-export declare function getImageTensor(input: tf.Tensor | NetInput): tf.Tensor4D;

build/commons/getImageTensor.js

-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-var tf = require("@tensorflow/tfjs-core");
-var NetInput_1 = require("../NetInput");
-function getImageTensor(input) {
-    return tf.tidy(function () {
-        if (input instanceof tf.Tensor) {
-            var rank = input.shape.length;
-            if (rank !== 3 && rank !== 4) {
-                throw new Error('input tensor must be of rank 3 or 4');
-            }
-            return (rank === 3 ? input.expandDims(0) : input).toFloat();
-        }
-        if (!(input instanceof NetInput_1.NetInput)) {
-            throw new Error('getImageTensor - expected input to be a tensor or an instance of NetInput');
-        }
-        return tf.concat(input.canvases.map(function (canvas) {
-            return tf.fromPixels(canvas).expandDims(0).toFloat();
-        }));
-    });
-}
-exports.getImageTensor = getImageTensor;
-//# sourceMappingURL=getImageTensor.js.map
\ No newline at end of file

build/commons/getImageTensor.js.map

-{"version":3,"file":"getImageTensor.js","sourceRoot":"","sources":["../../src/commons/getImageTensor.ts"],"names":[],"mappings":";;AAAA,0CAA4C;AAE5C,wCAAuC;AAEvC,wBAA+B,KAA2B;IACxD,OAAO,EAAE,CAAC,IAAI,CAAC;QACb,IAAI,KAAK,YAAY,EAAE,CAAC,MAAM,EAAE;YAC9B,IAAM,IAAI,GAAG,KAAK,CAAC,KAAK,CAAC,MAAM,CAAA;YAC/B,IAAI,IAAI,KAAK,CAAC,IAAI,IAAI,KAAK,CAAC,EAAE;gBAC5B,MAAM,IAAI,KAAK,CAAC,qCAAqC,CAAC,CAAA;aACvD;YAED,OAAO,CAAC,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,OAAO,EAAiB,CAAA;SAC3E;QAED,IAAI,CAAC,CAAC,KAAK,YAAY,mBAAQ,CAAC,EAAE;YAChC,MAAM,IAAI,KAAK,CAAC,2EAA2E,CAAC,CAAA;SAC7F;QAED,OAAO,EAAE,CAAC,MAAM,CACd,KAAK,CAAC,QAAQ,CAAC,GAAG,CAAC,UAAA,MAAM;YACvB,OAAA,EAAE,CAAC,UAAU,CAAC,MAAM,CAAC,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC,OAAO,EAAE;QAA7C,CAA6C,CAC9C,CACa,CAAA;IAClB,CAAC,CAAC,CAAA;AACJ,CAAC;AArBD,wCAqBC"}
\ No newline at end of file

build/commons/isMediaElement.d.ts

+export declare function isMediaElement(input: any): boolean;

build/commons/isMediaElement.js

+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+function isMediaElement(input) {
+    return input instanceof HTMLImageElement
+        || input instanceof HTMLVideoElement
+        || input instanceof HTMLCanvasElement;
+}
+exports.isMediaElement = isMediaElement;
+//# sourceMappingURL=isMediaElement.js.map
\ No newline at end of file

build/commons/isMediaElement.js.map

+{"version":3,"file":"isMediaElement.js","sourceRoot":"","sources":["../../src/commons/isMediaElement.ts"],"names":[],"mappings":";;AAAA,wBAA+B,KAAU;IACvC,OAAO,KAAK,YAAY,gBAAgB;WACnC,KAAK,YAAY,gBAAgB;WACjC,KAAK,YAAY,iBAAiB,CAAA;AACzC,CAAC;AAJD,wCAIC"}
\ No newline at end of file

build/commons/isTensor.d.ts

-import * as tf from '@tensorflow/tfjs-core';
-export declare function isTensor(tensor: tf.Tensor, dim: number): boolean;
-export declare function isTensor1D(tensor: tf.Tensor): boolean;
-export declare function isTensor2D(tensor: tf.Tensor): boolean;
-export declare function isTensor3D(tensor: tf.Tensor): boolean;
-export declare function isTensor4D(tensor: tf.Tensor): boolean;
+export declare function isTensor(tensor: any, dim: number): boolean;
+export declare function isTensor1D(tensor: any): boolean;
+export declare function isTensor2D(tensor: any): boolean;
+export declare function isTensor3D(tensor: any): boolean;
+export declare function isTensor4D(tensor: any): boolean;

build/commons/isTensor.js.map

-{"version":3,"file":"isTensor.js","sourceRoot":"","sources":["../../src/commons/isTensor.ts"],"names":[],"mappings":";;AAAA,0CAA4C;AAE5C,kBAAyB,MAAiB,EAAE,GAAW;IACrD,OAAO,MAAM,YAAY,EAAE,CAAC,MAAM,IAAI,MAAM,CAAC,KAAK,CAAC,MAAM,KAAK,GAAG,CAAA;AACnE,CAAC;AAFD,4BAEC;AAED,oBAA2B,MAAiB;IAC1C,OAAO,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,CAAA;AAC5B,CAAC;AAFD,gCAEC;AAED,oBAA2B,MAAiB;IAC1C,OAAO,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,CAAA;AAC5B,CAAC;AAFD,gCAEC;AAED,oBAA2B,MAAiB;IAC1C,OAAO,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,CAAA;AAC5B,CAAC;AAFD,gCAEC;AAED,oBAA2B,MAAiB;IAC1C,OAAO,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,CAAA;AAC5B,CAAC;AAFD,gCAEC"}
\ No newline at end of file
+{"version":3,"file":"isTensor.js","sourceRoot":"","sources":["../../src/commons/isTensor.ts"],"names":[],"mappings":";;AAAA,0CAA4C;AAE5C,kBAAyB,MAAW,EAAE,GAAW;IAC/C,OAAO,MAAM,YAAY,EAAE,CAAC,MAAM,IAAI,MAAM,CAAC,KAAK,CAAC,MAAM,KAAK,GAAG,CAAA;AACnE,CAAC;AAFD,4BAEC;AAED,oBAA2B,MAAW;IACpC,OAAO,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,CAAA;AAC5B,CAAC;AAFD,gCAEC;AAED,oBAA2B,MAAW;IACpC,OAAO,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,CAAA;AAC5B,CAAC;AAFD,gCAEC;AAED,oBAA2B,MAAW;IACpC,OAAO,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,CAAA;AAC5B,CAAC;AAFD,gCAEC;AAED,oBAA2B,MAAW;IACpC,OAAO,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,CAAA;AAC5B,CAAC;AAFD,gCAEC"}
\ No newline at end of file

build/commons/types.d.ts

@@ -4,3 +4,9 @@ export declare type ConvParams = {
     bias: tf.Tensor1D;
 };
 export declare type ExtractWeightsFunction = (numWeights: number) => Float32Array;
+export declare type BatchReshapeInfo = {
+    originalWidth: number;
+    originalHeight: number;
+    paddingX: number;
+    paddingY: number;
+};

build/drawing/index.js

@@ -26,7 +26,7 @@ function drawText(ctx, x, y, text, options) {
 }
 exports.drawText = drawText;
 function drawDetection(canvasArg, detection, options) {
-    var canvas = utils_1.getElement(canvasArg);
+    var canvas = utils_1.resolveInput(canvasArg);
     if (!(canvas instanceof HTMLCanvasElement)) {
         throw new Error('drawBox - expected canvas to be of type: HTMLCanvasElement');
     }
@@ -66,7 +66,7 @@ function drawContour(ctx, points, isClosed) {
     ctx.stroke();
 }
 function drawLandmarks(canvasArg, faceLandmarks, options) {
-    var canvas = utils_1.getElement(canvasArg);
+    var canvas = utils_1.resolveInput(canvasArg);
     if (!(canvas instanceof HTMLCanvasElement)) {
         throw new Error('drawLandmarks - expected canvas to be of type: HTMLCanvasElement');
     }

build/drawing/index.js.map

-{"version":3,"file":"index.js","sourceRoot":"","sources":["../../src/drawing/index.ts"],"names":[],"mappings":";;AAGA,kCAAkE;AAGlE;IACE,OAAO;QACL,KAAK,EAAE,MAAM;QACb,SAAS,EAAE,CAAC;QACZ,QAAQ,EAAE,EAAE;QACZ,SAAS,EAAE,SAAS;KACrB,CAAA;AACH,CAAC;AAPD,sDAOC;AAED,iBACE,GAA6B,EAC7B,CAAS,EACT,CAAS,EACT,CAAS,EACT,CAAS,EACT,OAAuB;IAEvB,IAAM,WAAW,GAAG,MAAM,CAAC,MAAM,CAC/B,qBAAqB,EAAE,EACvB,CAAC,OAAO,IAAI,EAAE,CAAC,CAChB,CAAA;IAED,GAAG,CAAC,WAAW,GAAG,WAAW,CAAC,KAAK,CAAA;IACnC,GAAG,CAAC,SAAS,GAAG,WAAW,CAAC,SAAS,CAAA;IACrC,GAAG,CAAC,UAAU,CAAC,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC,CAAA;AAC5B,CAAC;AAhBD,0BAgBC;AAED,kBACE,GAA6B,EAC7B,CAAS,EACT,CAAS,EACT,IAAY,EACZ,OAAwB;IAExB,IAAM,WAAW,GAAG,MAAM,CAAC,MAAM,CAC/B,qBAAqB,EAAE,EACvB,CAAC,OAAO,IAAI,EAAE,CAAC,CAChB,CAAA;IAED,IAAM,OAAO,GAAG,CAAC,GAAG,WAAW,CAAC,SAAS,CAAA;IAEzC,GAAG,CAAC,SAAS,GAAG,WAAW,CAAC,KAAK,CAAA;IACjC,GAAG,CAAC,IAAI,GAAM,WAAW,CAAC,QAAQ,WAAM,WAAW,CAAC,SAAW,CAAA;IAC/D,GAAG,CAAC,QAAQ,CAAC,IAAI,EAAE,CAAC,GAAG,OAAO,EAAE,CAAC,GAAG,OAAO,GAAG,CAAC,WAAW,CAAC,QAAQ,GAAG,GAAG,CAAC,CAAC,CAAA;AAC7E,CAAC;AAjBD,4BAiBC;AAED,uBACE,SAAqC,EACrC,SAA0C,EAC1C,OAAmE;IAEnE,IAAM,MAAM,GAAG,kBAAU,CAAC,SAAS,CAAC,CAAA;IACpC,IAAI,CAAC,CAAC,MAAM,YAAY,iBAAiB,CAAC,EAAE;QAC1C,MAAM,IAAI,KAAK,CAAC,4DAA4D,CAAC,CAAA;KAC9E;IAED,IAAM,cAAc,GAAG,KAAK,CAAC,OAAO,CAAC,SAAS,CAAC;QAC7C,CAAC,CAAC,SAAS;QACX,CAAC,CAAC,CAAC,SAAS,CAAC,CAAA;IAEf,cAAc,CAAC,OAAO,CAAC,UAAC,GAAG;QACnB,IAAA,iBAKU,EAJd,QAAC,EACD,QAAC,EACD,gBAAK,EACL,kBAAM,CACQ;QAEhB,IAAM,WAAW,GAAG,MAAM,CAAC,MAAM,CAC/B,qBAAqB,EAAE,EACvB,CAAC,OAAO,IAAI,EAAE,CAAC,CAChB,CAAA;QAEO,IAAA,yEAAS,CAAwD;QAEzE,IAAM,GAAG,GAAG,2BAAmB,CAAC,MAAM,CAAC,CAAA;QACvC,OAAO,CACL,GAAG,EACH,CAAC,EACD,CAAC,EACD,KAAK,EACL,MAAM,EACN,WAAW,CACZ,CAAA;QACD,IAAI,SAAS,EAAE;YACb,QAAQ,CACN,GAAG,EACH,CAAC,EACD,CAAC,EACD,KAAG,aAAK,CAAC,GAAG,CAAC,QAAQ,EAAE,CAAG,EAC1B,WAAW,CACZ,CAAA;SACF;IACH,CAAC,CAAC,CAAA;AACJ,CAAC;AAhDD,sCAgDC;AAED,qBACE,GAA6B,EAC7B,MAAe,EACf,QAAyB;IAAzB,yBAAA,EAAA,gBAAyB;IAEzB,GAAG,CAAC,SAAS,EAAE,CAAA;IAEf,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,UAAC,EAAQ,EAAE,OAAO;YAAf,QAAC,EAAE,QAAC;QAC7B,IAAM,IAAI,GAAG,MAAM,CAAC,OAAO,CAAC,CAAA;QAC5B,GAAG,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,EAAE,IAAI,CAAC,CAAC,CAAC,CAAA;QAC1B,GAAG,CAAC,MAAM,CAAC,CAAC,EAAE,CAAC,CAAC,CAAA;IAClB,CAAC,CAAC,CAAA;IAEF,IAAI,QAAQ,EAAE;QACZ,IAAM,IAAI,GAAG,MAAM,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,CAAA;QACtC,IAAM,EAAE,GAAG,MAAM,CAAC,CAAC,CAAC,CAAA;QACpB,IAAI,CAAC,IAAI,IAAI,CAAC,EAAE,EAAE;YAChB,OAAM;SACP;QAED,GAAG,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,EAAE,IAAI,CAAC,CAAC,CAAC,CAAA;QAC1B,GAAG,CAAC,MAAM,CAAC,EAAE,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAA;KACvB;IAED,GAAG,CAAC,MAAM,EAAE,CAAA;AACd,CAAC;AAED,uBACE,SAAqC,EACrC,aAA8C,EAC9C,OAAuD;IAEvD,IAAM,MAAM,GAAG,kBAAU,CAAC,SAAS,CAAC,CAAA;IACpC,IAAI,CAAC,CAAC,MAAM,YAAY,iBAAiB,CAAC,EAAE;QAC1C,MAAM,IAAI,KAAK,CAAC,kEAAkE,CAAC,CAAA;KACpF;IAED,IAAM,WAAW,GAAG,MAAM,CAAC,MAAM,CAC/B,qBAAqB,EAAE,EACvB,CAAC,OAAO,IAAI,EAAE,CAAC,CAChB,CAAA;IAEO,IAAA,0EAAS,CAAyD;IAE1E,IAAM,GAAG,GAAG,2BAAmB,CAAC,MAAM,CAAC,CAAA;IAC/B,IAAA,iCAAS,EAAE,yBAAK,CAAgB;IAExC,IAAM,kBAAkB,GAAG,KAAK,CAAC,OAAO,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,aAAa,CAAC,CAAA;IAEzF,kBAAkB,CAAC,OAAO,CAAC,UAAA,SAAS;QAClC,IAAI,SAAS,EAAE;YACb,GAAG,CAAC,WAAW,GAAG,KAAK,CAAA;YACvB,GAAG,CAAC,SAAS,GAAG,SAAS,CAAA;YACzB,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,aAAa,EAAE,CAAC,CAAA;YAC3C,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,cAAc,EAAE,CAAC,CAAA;YAC5C,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,eAAe,EAAE,CAAC,CAAA;YAC7C,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,OAAO,EAAE,CAAC,CAAA;YACrC,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,UAAU,EAAE,EAAE,IAAI,CAAC,CAAA;YAC9C,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,WAAW,EAAE,EAAE,IAAI,CAAC,CAAA;YAC/C,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,QAAQ,EAAE,EAAE,IAAI,CAAC,CAAA;YAC5C,OAAM;SACP;QAED,mBAAmB;QACnB,IAAM,QAAQ,GAAG,SAAS,GAAG,CAAC,CAAA;QAC9B,GAAG,CAAC,SAAS,GAAG,KAAK,CAAA;QACrB,SAAS,CAAC,YAAY,EAAE,CAAC,OAAO,CAAC,UAAA,EAAE,IAAI,OAAA,GAAG,CAAC,QAAQ,CAAC,EAAE,CAAC,CAAC,GAAG,QAAQ,EAAE,EAAE,CAAC,CAAC,GAAG,QAAQ,EAAE,SAAS,EAAE,SAAS,CAAC,EAApE,CAAoE,CAAC,CAAA;IAC9G,CAAC,CAAC,CAAA;AACJ,CAAC;AAzCD,sCAyCC"}
\ No newline at end of file
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../../src/drawing/index.ts"],"names":[],"mappings":";;AAGA,kCAAoE;AAGpE;IACE,OAAO;QACL,KAAK,EAAE,MAAM;QACb,SAAS,EAAE,CAAC;QACZ,QAAQ,EAAE,EAAE;QACZ,SAAS,EAAE,SAAS;KACrB,CAAA;AACH,CAAC;AAPD,sDAOC;AAED,iBACE,GAA6B,EAC7B,CAAS,EACT,CAAS,EACT,CAAS,EACT,CAAS,EACT,OAAuB;IAEvB,IAAM,WAAW,GAAG,MAAM,CAAC,MAAM,CAC/B,qBAAqB,EAAE,EACvB,CAAC,OAAO,IAAI,EAAE,CAAC,CAChB,CAAA;IAED,GAAG,CAAC,WAAW,GAAG,WAAW,CAAC,KAAK,CAAA;IACnC,GAAG,CAAC,SAAS,GAAG,WAAW,CAAC,SAAS,CAAA;IACrC,GAAG,CAAC,UAAU,CAAC,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC,CAAA;AAC5B,CAAC;AAhBD,0BAgBC;AAED,kBACE,GAA6B,EAC7B,CAAS,EACT,CAAS,EACT,IAAY,EACZ,OAAwB;IAExB,IAAM,WAAW,GAAG,MAAM,CAAC,MAAM,CAC/B,qBAAqB,EAAE,EACvB,CAAC,OAAO,IAAI,EAAE,CAAC,CAChB,CAAA;IAED,IAAM,OAAO,GAAG,CAAC,GAAG,WAAW,CAAC,SAAS,CAAA;IAEzC,GAAG,CAAC,SAAS,GAAG,WAAW,CAAC,KAAK,CAAA;IACjC,GAAG,CAAC,IAAI,GAAM,WAAW,CAAC,QAAQ,WAAM,WAAW,CAAC,SAAW,CAAA;IAC/D,GAAG,CAAC,QAAQ,CAAC,IAAI,EAAE,CAAC,GAAG,OAAO,EAAE,CAAC,GAAG,OAAO,GAAG,CAAC,WAAW,CAAC,QAAQ,GAAG,GAAG,CAAC,CAAC,CAAA;AAC7E,CAAC;AAjBD,4BAiBC;AAED,uBACE,SAAqC,EACrC,SAA0C,EAC1C,OAAmE;IAEnE,IAAM,MAAM,GAAG,oBAAY,CAAC,SAAS,CAAC,CAAA;IACtC,IAAI,CAAC,CAAC,MAAM,YAAY,iBAAiB,CAAC,EAAE;QAC1C,MAAM,IAAI,KAAK,CAAC,4DAA4D,CAAC,CAAA;KAC9E;IAED,IAAM,cAAc,GAAG,KAAK,CAAC,OAAO,CAAC,SAAS,CAAC;QAC7C,CAAC,CAAC,SAAS;QACX,CAAC,CAAC,CAAC,SAAS,CAAC,CAAA;IAEf,cAAc,CAAC,OAAO,CAAC,UAAC,GAAG;QACnB,IAAA,iBAKU,EAJd,QAAC,EACD,QAAC,EACD,gBAAK,EACL,kBAAM,CACQ;QAEhB,IAAM,WAAW,GAAG,MAAM,CAAC,MAAM,CAC/B,qBAAqB,EAAE,EACvB,CAAC,OAAO,IAAI,EAAE,CAAC,CAChB,CAAA;QAEO,IAAA,yEAAS,CAAwD;QAEzE,IAAM,GAAG,GAAG,2BAAmB,CAAC,MAAM,CAAC,CAAA;QACvC,OAAO,CACL,GAAG,EACH,CAAC,EACD,CAAC,EACD,KAAK,EACL,MAAM,EACN,WAAW,CACZ,CAAA;QACD,IAAI,SAAS,EAAE;YACb,QAAQ,CACN,GAAG,EACH,CAAC,EACD,CAAC,EACD,KAAG,aAAK,CAAC,GAAG,CAAC,QAAQ,EAAE,CAAG,EAC1B,WAAW,CACZ,CAAA;SACF;IACH,CAAC,CAAC,CAAA;AACJ,CAAC;AAhDD,sCAgDC;AAED,qBACE,GAA6B,EAC7B,MAAe,EACf,QAAyB;IAAzB,yBAAA,EAAA,gBAAyB;IAEzB,GAAG,CAAC,SAAS,EAAE,CAAA;IAEf,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,UAAC,EAAQ,EAAE,OAAO;YAAf,QAAC,EAAE,QAAC;QAC7B,IAAM,IAAI,GAAG,MAAM,CAAC,OAAO,CAAC,CAAA;QAC5B,GAAG,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,EAAE,IAAI,CAAC,CAAC,CAAC,CAAA;QAC1B,GAAG,CAAC,MAAM,CAAC,CAAC,EAAE,CAAC,CAAC,CAAA;IAClB,CAAC,CAAC,CAAA;IAEF,IAAI,QAAQ,EAAE;QACZ,IAAM,IAAI,GAAG,MAAM,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,CAAA;QACtC,IAAM,EAAE,GAAG,MAAM,CAAC,CAAC,CAAC,CAAA;QACpB,IAAI,CAAC,IAAI,IAAI,CAAC,EAAE,EAAE;YAChB,OAAM;SACP;QAED,GAAG,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,EAAE,IAAI,CAAC,CAAC,CAAC,CAAA;QAC1B,GAAG,CAAC,MAAM,CAAC,EAAE,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAA;KACvB;IAED,GAAG,CAAC,MAAM,EAAE,CAAA;AACd,CAAC;AAED,uBACE,SAAqC,EACrC,aAA8C,EAC9C,OAAuD;IAEvD,IAAM,MAAM,GAAG,oBAAY,CAAC,SAAS,CAAC,CAAA;IACtC,IAAI,CAAC,CAAC,MAAM,YAAY,iBAAiB,CAAC,EAAE;QAC1C,MAAM,IAAI,KAAK,CAAC,kEAAkE,CAAC,CAAA;KACpF;IAED,IAAM,WAAW,GAAG,MAAM,CAAC,MAAM,CAC/B,qBAAqB,EAAE,EACvB,CAAC,OAAO,IAAI,EAAE,CAAC,CAChB,CAAA;IAEO,IAAA,0EAAS,CAAyD;IAE1E,IAAM,GAAG,GAAG,2BAAmB,CAAC,MAAM,CAAC,CAAA;IAC/B,IAAA,iCAAS,EAAE,yBAAK,CAAgB;IAExC,IAAM,kBAAkB,GAAG,KAAK,CAAC,OAAO,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,aAAa,CAAC,CAAA;IAEzF,kBAAkB,CAAC,OAAO,CAAC,UAAA,SAAS;QAClC,IAAI,SAAS,EAAE;YACb,GAAG,CAAC,WAAW,GAAG,KAAK,CAAA;YACvB,GAAG,CAAC,SAAS,GAAG,SAAS,CAAA;YACzB,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,aAAa,EAAE,CAAC,CAAA;YAC3C,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,cAAc,EAAE,CAAC,CAAA;YAC5C,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,eAAe,EAAE,CAAC,CAAA;YAC7C,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,OAAO,EAAE,CAAC,CAAA;YACrC,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,UAAU,EAAE,EAAE,IAAI,CAAC,CAAA;YAC9C,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,WAAW,EAAE,EAAE,IAAI,CAAC,CAAA;YAC/C,WAAW,CAAC,GAAG,EAAE,SAAS,CAAC,QAAQ,EAAE,EAAE,IAAI,CAAC,CAAA;YAC5C,OAAM;SACP;QAED,mBAAmB;QACnB,IAAM,QAAQ,GAAG,SAAS,GAAG,CAAC,CAAA;QAC9B,GAAG,CAAC,SAAS,GAAG,KAAK,CAAA;QACrB,SAAS,CAAC,YAAY,EAAE,CAAC,OAAO,CAAC,UAAA,EAAE,IAAI,OAAA,GAAG,CAAC,QAAQ,CAAC,EAAE,CAAC,CAAC,GAAG,QAAQ,EAAE,EAAE,CAAC,CAAC,GAAG,QAAQ,EAAE,SAAS,EAAE,SAAS,CAAC,EAApE,CAAoE,CAAC,CAAA;IAC9G,CAAC,CAAC,CAAA;AACJ,CAAC;AAzCD,sCAyCC"}
\ No newline at end of file

build/extractFaceTensors.d.ts

 import * as tf from '@tensorflow/tfjs-core';
 import { FaceDetection } from './faceDetectionNet/FaceDetection';
-import { NetInput } from './NetInput';
 import { Rect } from './Rect';
 import { TNetInput } from './types';
 /**
@@ -13,4 +12,4 @@ import { TNetInput } from './types';
  * @param detections The face detection results or face bounding boxes for that image.
  * @returns Tensors of the corresponding image region for each detected face.
  */
-export declare function extractFaceTensors(input: tf.Tensor | NetInput | TNetInput, detections: Array<FaceDetection | Rect>): Promise<tf.Tensor4D[]>;
+export declare function extractFaceTensors(input: TNetInput, detections: Array<FaceDetection | Rect>): Promise<tf.Tensor4D[]>;

build/extractFaceTensors.js

@@ -2,7 +2,6 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 var tslib_1 = require("tslib");
 var tf = require("@tensorflow/tfjs-core");
-var getImageTensor_1 = require("./commons/getImageTensor");
 var FaceDetection_1 = require("./faceDetectionNet/FaceDetection");
 var toNetInput_1 = require("./toNetInput");
 /**
@@ -17,23 +16,21 @@ var toNetInput_1 = require("./toNetInput");
  */
 function extractFaceTensors(input, detections) {
     return tslib_1.__awaiter(this, void 0, void 0, function () {
-        var image, _a;
-        return tslib_1.__generator(this, function (_b) {
-            switch (_b.label) {
-                case 0:
-                    if (!(input instanceof tf.Tensor)) return [3 /*break*/, 1];
-                    _a = input;
-                    return [3 /*break*/, 3];
-                case 1: return [4 /*yield*/, toNetInput_1.toNetInput(input)];
-                case 2:
-                    _a = _b.sent();
-                    _b.label = 3;
-                case 3:
-                    image = _a;
+        var netInput;
+        return tslib_1.__generator(this, function (_a) {
+            switch (_a.label) {
+                case 0: return [4 /*yield*/, toNetInput_1.toNetInput(input, true)];
+                case 1:
+                    netInput = _a.sent();
+                    if (netInput.batchSize > 1) {
+                        if (netInput.isManaged) {
+                            netInput.dispose();
+                        }
+                        throw new Error('extractFaceTensors - batchSize > 1 not supported');
+                    }
                     return [2 /*return*/, tf.tidy(function () {
-                            var imgTensor = getImageTensor_1.getImageTensor(image);
-                            // TODO handle batches
-                            var _a = imgTensor.shape, batchSize = _a[0], imgHeight = _a[1], imgWidth = _a[2], numChannels = _a[3];
+                            var imgTensor = netInput.inputs[0].expandDims().toFloat();
+                            var _a = imgTensor.shape.slice(1), imgHeight = _a[0], imgWidth = _a[1], numChannels = _a[2];
                             var boxes = detections.map(function (det) { return det instanceof FaceDetection_1.FaceDetection
                                 ? det.forSize(imgWidth, imgHeight).getBox().floor()
                                 : det; });
@@ -41,6 +38,9 @@ function extractFaceTensors(input, detections) {
                                 var x = _a.x, y = _a.y, width = _a.width, height = _a.height;
                                 return tf.slice(imgTensor, [0, y, x, 0], [1, height, width, numChannels]);
                             });
+                            if (netInput.isManaged) {
+                                netInput.dispose();
+                            }
                             return faceTensors;
                         })];
             }

build/extractFaceTensors.js.map

-{"version":3,"file":"extractFaceTensors.js","sourceRoot":"","sources":["../src/extractFaceTensors.ts"],"names":[],"mappings":";;;AAAA,0CAA4C;AAE5C,2DAA0D;AAC1D,kEAAiE;AAIjE,2CAA0C;AAE1C;;;;;;;;;GASG;AACH,4BACE,KAAuC,EACvC,UAAqC;;;;;;yBAGvB,CAAA,KAAK,YAAY,EAAE,CAAC,MAAM,CAAA,EAA1B,wBAA0B;oBACpC,KAAA,KAAK,CAAA;;wBACL,qBAAM,uBAAU,CAAC,KAAK,CAAC,EAAA;;oBAAvB,KAAA,SAAuB,CAAA;;;oBAFrB,KAAK,KAEgB;oBAE3B,sBAAO,EAAE,CAAC,IAAI,CAAC;4BACb,IAAM,SAAS,GAAG,+BAAc,CAAC,KAAK,CAAC,CAAA;4BAEvC,sBAAsB;4BAChB,IAAA,oBAA+D,EAA9D,iBAAS,EAAE,iBAAS,EAAE,gBAAQ,EAAE,mBAAW,CAAmB;4BAErE,IAAM,KAAK,GAAG,UAAU,CAAC,GAAG,CAC1B,UAAA,GAAG,IAAI,OAAA,GAAG,YAAY,6BAAa;gCACjC,CAAC,CAAC,GAAG,CAAC,OAAO,CAAC,QAAQ,EAAE,SAAS,CAAC,CAAC,MAAM,EAAE,CAAC,KAAK,EAAE;gCACnD,CAAC,CAAC,GAAG,EAFA,CAEA,CACR,CAAA;4BACD,IAAM,WAAW,GAAG,KAAK,CAAC,GAAG,CAAC,UAAC,EAAuB;oCAArB,QAAC,EAAE,QAAC,EAAE,gBAAK,EAAE,kBAAM;gCAClD,OAAA,EAAE,CAAC,KAAK,CAAC,SAAS,EAAE,CAAC,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,EAAE,WAAW,CAAC,CAAC;4BAAlE,CAAkE,CACnE,CAAA;4BAED,OAAO,WAAW,CAAA;wBACpB,CAAC,CAAC,EAAA;;;;CACH;AA1BD,gDA0BC"}
\ No newline at end of file
+{"version":3,"file":"extractFaceTensors.js","sourceRoot":"","sources":["../src/extractFaceTensors.ts"],"names":[],"mappings":";;;AAAA,0CAA4C;AAE5C,kEAAiE;AAEjE,2CAA0C;AAG1C;;;;;;;;;GASG;AACH,4BACE,KAAgB,EAChB,UAAuC;;;;;wBAGtB,qBAAM,uBAAU,CAAC,KAAK,EAAE,IAAI,CAAC,EAAA;;oBAAxC,QAAQ,GAAG,SAA6B;oBAE9C,IAAI,QAAQ,CAAC,SAAS,GAAG,CAAC,EAAE;wBAC1B,IAAI,QAAQ,CAAC,SAAS,EAAE;4BACtB,QAAQ,CAAC,OAAO,EAAE,CAAA;yBACnB;wBACD,MAAM,IAAI,KAAK,CAAC,kDAAkD,CAAC,CAAA;qBACpE;oBAED,sBAAO,EAAE,CAAC,IAAI,CAAC;4BACb,IAAM,SAAS,GAAG,QAAQ,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,UAAU,EAAE,CAAC,OAAO,EAAiB,CAAA;4BAEpE,IAAA,6BAA6D,EAA5D,iBAAS,EAAE,gBAAQ,EAAE,mBAAW,CAA4B;4BAEnE,IAAM,KAAK,GAAG,UAAU,CAAC,GAAG,CAC1B,UAAA,GAAG,IAAI,OAAA,GAAG,YAAY,6BAAa;gCACjC,CAAC,CAAC,GAAG,CAAC,OAAO,CAAC,QAAQ,EAAE,SAAS,CAAC,CAAC,MAAM,EAAE,CAAC,KAAK,EAAE;gCACnD,CAAC,CAAC,GAAG,EAFA,CAEA,CACR,CAAA;4BACD,IAAM,WAAW,GAAG,KAAK,CAAC,GAAG,CAAC,UAAC,EAAuB;oCAArB,QAAC,EAAE,QAAC,EAAE,gBAAK,EAAE,kBAAM;gCAClD,OAAA,EAAE,CAAC,KAAK,CAAC,SAAS,EAAE,CAAC,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,EAAE,WAAW,CAAC,CAAC;4BAAlE,CAAkE,CACnE,CAAA;4BAED,IAAI,QAAQ,CAAC,SAAS,EAAE;gCACtB,QAAQ,CAAC,OAAO,EAAE,CAAA;6BACnB;4BACD,OAAO,WAAW,CAAA;wBACpB,CAAC,CAAC,EAAA;;;;CACH;AAjCD,gDAiCC"}
\ No newline at end of file

build/extractFaces.d.ts

 import { FaceDetection } from './faceDetectionNet/FaceDetection';
 import { Rect } from './Rect';
+import { TNetInput } from './types';
 /**
  * Extracts the image regions containing the detected faces.
  *
@@ -7,4 +8,4 @@ import { Rect } from './Rect';
  * @param detections The face detection results or face bounding boxes for that image.
  * @returns The Canvases of the corresponding image region for each detected face.
  */
-export declare function extractFaces(image: HTMLCanvasElement, detections: Array<FaceDetection | Rect>): HTMLCanvasElement[];
+export declare function extractFaces(input: TNetInput, detections: Array<FaceDetection | Rect>): Promise<HTMLCanvasElement[]>;

build/extractFaces.js

 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
+var tslib_1 = require("tslib");
 var FaceDetection_1 = require("./faceDetectionNet/FaceDetection");
+var toNetInput_1 = require("./toNetInput");
 var utils_1 = require("./utils");
 /**
  * Extracts the image regions containing the detected faces.
@@ -9,17 +11,41 @@ var utils_1 = require("./utils");
  * @param detections The face detection results or face bounding boxes for that image.
  * @returns The Canvases of the corresponding image region for each detected face.
  */
-function extractFaces(image, detections) {
-    var ctx = utils_1.getContext2dOrThrow(image);
-    var boxes = detections.map(function (det) { return det instanceof FaceDetection_1.FaceDetection
-        ? det.forSize(image.width, image.height).getBox().floor()
-        : det; });
-    return boxes.map(function (_a) {
-        var x = _a.x, y = _a.y, width = _a.width, height = _a.height;
-        var faceImg = utils_1.createCanvas({ width: width, height: height });
-        utils_1.getContext2dOrThrow(faceImg)
-            .putImageData(ctx.getImageData(x, y, width, height), 0, 0);
-        return faceImg;
+function extractFaces(input, detections) {
+    return tslib_1.__awaiter(this, void 0, void 0, function () {
+        var canvas, netInput, ctx, boxes;
+        return tslib_1.__generator(this, function (_a) {
+            switch (_a.label) {
+                case 0:
+                    canvas = input;
+                    if (!!(input instanceof HTMLCanvasElement)) return [3 /*break*/, 3];
+                    return [4 /*yield*/, toNetInput_1.toNetInput(input, true)];
+                case 1:
+                    netInput = _a.sent();
+                    if (netInput.batchSize > 1) {
+                        if (netInput.isManaged) {
+                            netInput.dispose();
+                        }
+                        throw new Error('extractFaces - batchSize > 1 not supported');
+                    }
+                    return [4 /*yield*/, utils_1.imageTensorToCanvas(netInput.inputs[0])];
+                case 2:
+                    canvas = _a.sent();
+                    _a.label = 3;
+                case 3:
+                    ctx = utils_1.getContext2dOrThrow(canvas);
+                    boxes = detections.map(function (det) { return det instanceof FaceDetection_1.FaceDetection
+                        ? det.forSize(canvas.width, canvas.height).getBox().floor()
+                        : det; });
+                    return [2 /*return*/, boxes.map(function (_a) {
+                            var x = _a.x, y = _a.y, width = _a.width, height = _a.height;
+                            var faceImg = utils_1.createCanvas({ width: width, height: height });
+                            utils_1.getContext2dOrThrow(faceImg)
+                                .putImageData(ctx.getImageData(x, y, width, height), 0, 0);
+                            return faceImg;
+                        })];
+            }
+        });
     });
 }
 exports.extractFaces = extractFaces;

build/extractFaces.js.map

-{"version":3,"file":"extractFaces.js","sourceRoot":"","sources":["../src/extractFaces.ts"],"names":[],"mappings":";;AAAA,kEAAiE;AAEjE,iCAA4D;AAE5D;;;;;;GAMG;AACH,sBACE,KAAwB,EACxB,UAAqC;IAErC,IAAM,GAAG,GAAG,2BAAmB,CAAC,KAAK,CAAC,CAAA;IAEtC,IAAM,KAAK,GAAG,UAAU,CAAC,GAAG,CAC1B,UAAA,GAAG,IAAI,OAAA,GAAG,YAAY,6BAAa;QACjC,CAAC,CAAC,GAAG,CAAC,OAAO,CAAC,KAAK,CAAC,KAAK,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC,MAAM,EAAE,CAAC,KAAK,EAAE;QACzD,CAAC,CAAC,GAAG,EAFA,CAEA,CACR,CAAA;IACD,OAAO,KAAK,CAAC,GAAG,CAAC,UAAC,EAAuB;YAArB,QAAC,EAAE,QAAC,EAAE,gBAAK,EAAE,kBAAM;QACrC,IAAM,OAAO,GAAG,oBAAY,CAAC,EAAE,KAAK,OAAA,EAAE,MAAM,QAAA,EAAE,CAAC,CAAA;QAC/C,2BAAmB,CAAC,OAAO,CAAC;aACzB,YAAY,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC,EAAE,CAAC,EAAE,KAAK,EAAE,MAAM,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC,CAAA;QAC5D,OAAO,OAAO,CAAA;IAChB,CAAC,CAAC,CAAA;AACJ,CAAC;AAjBD,oCAiBC"}
\ No newline at end of file
+{"version":3,"file":"extractFaces.js","sourceRoot":"","sources":["../src/extractFaces.ts"],"names":[],"mappings":";;;AAAA,kEAAiE;AAEjE,2CAA0C;AAE1C,iCAAiF;AAEjF;;;;;;GAMG;AACH,sBACE,KAAgB,EAChB,UAAuC;;;;;;oBAGnC,MAAM,GAAG,KAA0B,CAAA;yBAEnC,CAAC,CAAC,KAAK,YAAY,iBAAiB,CAAC,EAArC,wBAAqC;oBACtB,qBAAM,uBAAU,CAAC,KAAK,EAAE,IAAI,CAAC,EAAA;;oBAAxC,QAAQ,GAAG,SAA6B;oBAE9C,IAAI,QAAQ,CAAC,SAAS,GAAG,CAAC,EAAE;wBAC1B,IAAI,QAAQ,CAAC,SAAS,EAAE;4BACtB,QAAQ,CAAC,OAAO,EAAE,CAAA;yBACnB;wBACD,MAAM,IAAI,KAAK,CAAC,4CAA4C,CAAC,CAAA;qBAC9D;oBAEQ,qBAAM,2BAAmB,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,EAAA;;oBAAtD,MAAM,GAAG,SAA6C,CAAA;;;oBAGlD,GAAG,GAAG,2BAAmB,CAAC,MAAM,CAAC,CAAA;oBAEjC,KAAK,GAAG,UAAU,CAAC,GAAG,CAC1B,UAAA,GAAG,IAAI,OAAA,GAAG,YAAY,6BAAa;wBACjC,CAAC,CAAC,GAAG,CAAC,OAAO,CAAC,MAAM,CAAC,KAAK,EAAE,MAAM,CAAC,MAAM,CAAC,CAAC,MAAM,EAAE,CAAC,KAAK,EAAE;wBAC3D,CAAC,CAAC,GAAG,EAFA,CAEA,CACR,CAAA;oBACD,sBAAO,KAAK,CAAC,GAAG,CAAC,UAAC,EAAuB;gCAArB,QAAC,EAAE,QAAC,EAAE,gBAAK,EAAE,kBAAM;4BACrC,IAAM,OAAO,GAAG,oBAAY,CAAC,EAAE,KAAK,OAAA,EAAE,MAAM,QAAA,EAAE,CAAC,CAAA;4BAC/C,2BAAmB,CAAC,OAAO,CAAC;iCACzB,YAAY,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC,EAAE,CAAC,EAAE,KAAK,EAAE,MAAM,CAAC,EAAE,CAAC,EAAE,CAAC,CAAC,CAAA;4BAC5D,OAAO,OAAO,CAAA;wBAChB,CAAC,CAAC,EAAA;;;;CACH;AAjCD,oCAiCC"}
\ No newline at end of file

build/faceDetectionNet/FaceDetectionNet.d.ts

@@ -6,10 +6,13 @@ export declare class FaceDetectionNet {
     private _params;
     load(weightsOrUrl?: Float32Array | string): Promise<void>;
     extractWeights(weights: Float32Array): void;
-    private forwardTensor(imgTensor);
-    forward(input: tf.Tensor | NetInput | TNetInput): Promise<{
+    forwardInput(input: NetInput): {
+        boxes: tf.Tensor<tf.Rank.R2>[];
+        scores: tf.Tensor<tf.Rank.R1>[];
+    };
+    forward(input: TNetInput): Promise<{
         boxes: tf.Tensor<tf.Rank.R2>[];
         scores: tf.Tensor<tf.Rank.R1>[];
     }>;
-    locateFaces(input: tf.Tensor | NetInput | TNetInput, minConfidence?: number, maxResults?: number): Promise<FaceDetection[]>;
+    locateFaces(input: TNetInput, minConfidence?: number, maxResults?: number): Promise<FaceDetection[]>;
 }

build/faceDetectionNet/FaceDetectionNet.js

@@ -2,8 +2,6 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 var tslib_1 = require("tslib");
 var tf = require("@tensorflow/tfjs-core");
-var getImageTensor_1 = require("../commons/getImageTensor");
-var padToSquare_1 = require("../padToSquare");
 var Rect_1 = require("../Rect");
 var toNetInput_1 = require("../toNetInput");
 var extractParams_1 = require("./extractParams");
@@ -13,7 +11,6 @@ var mobileNetV1_1 = require("./mobileNetV1");
 var nonMaxSuppression_1 = require("./nonMaxSuppression");
 var outputLayer_1 = require("./outputLayer");
 var predictionLayer_1 = require("./predictionLayer");
-var resizeLayer_1 = require("./resizeLayer");
 var FaceDetectionNet = /** @class */ (function () {
     function FaceDetectionNet() {
     }
@@ -42,37 +39,28 @@ var FaceDetectionNet = /** @class */ (function () {
     FaceDetectionNet.prototype.extractWeights = function (weights) {
         this._params = extractParams_1.extractParams(weights);
     };
-    FaceDetectionNet.prototype.forwardTensor = function (imgTensor) {
+    FaceDetectionNet.prototype.forwardInput = function (input) {
         var _this = this;
         if (!this._params) {
             throw new Error('FaceDetectionNet - load model before inference');
         }
         return tf.tidy(function () {
-            var resized = resizeLayer_1.resizeLayer(imgTensor);
-            var features = mobileNetV1_1.mobileNetV1(resized, _this._params.mobilenetv1_params);
+            var batchTensor = input.toBatchTensor(512, false);
+            var x = tf.sub(tf.mul(batchTensor, tf.scalar(0.007843137718737125)), tf.scalar(1));
+            var features = mobileNetV1_1.mobileNetV1(x, _this._params.mobilenetv1_params);
             var _a = predictionLayer_1.predictionLayer(features.out, features.conv11, _this._params.prediction_layer_params), boxPredictions = _a.boxPredictions, classPredictions = _a.classPredictions;
             return outputLayer_1.outputLayer(boxPredictions, classPredictions, _this._params.output_layer_params);
         });
     };
     FaceDetectionNet.prototype.forward = function (input) {
         return tslib_1.__awaiter(this, void 0, void 0, function () {
-            var _this = this;
-            var netInput, _a;
+            var _a;
             return tslib_1.__generator(this, function (_b) {
                 switch (_b.label) {
                     case 0:
-                        if (!(input instanceof tf.Tensor)) return [3 /*break*/, 1];
-                        _a = input;
-                        return [3 /*break*/, 3];
-                    case 1: return [4 /*yield*/, toNetInput_1.toNetInput(input)];
-                    case 2:
-                        _a = _b.sent();
-                        _b.label = 3;
-                    case 3:
-                        netInput = _a;
-                        return [2 /*return*/, tf.tidy(function () {
-                                return _this.forwardTensor(padToSquare_1.padToSquare(getImageTensor_1.getImageTensor(netInput)));
-                            })];
+                        _a = this.forwardInput;
+                        return [4 /*yield*/, toNetInput_1.toNetInput(input, true)];
+                    case 1: return [2 /*return*/, _a.apply(this, [_b.sent()])];
                 }
             });
         });
@@ -81,42 +69,27 @@ var FaceDetectionNet = /** @class */ (function () {
         if (minConfidence === void 0) { minConfidence = 0.8; }
         if (maxResults === void 0) { maxResults = 100; }
         return tslib_1.__awaiter(this, void 0, void 0, function () {
-            var _this = this;
-            var netInput, _a, paddedHeightRelative, paddedWidthRelative, imageDimensions, _b, _boxes, _scores, boxes, scores, i, scoresData, _c, _d, iouThreshold, indices, results;
-            return tslib_1.__generator(this, function (_e) {
-                switch (_e.label) {
-                    case 0:
-                        if (!(input instanceof tf.Tensor)) return [3 /*break*/, 1];
-                        _a = input;
-                        return [3 /*break*/, 3];
-                    case 1: return [4 /*yield*/, toNetInput_1.toNetInput(input)];
-                    case 2:
-                        _a = _e.sent();
-                        _e.label = 3;
-                    case 3:
-                        netInput = _a;
-                        paddedHeightRelative = 1, paddedWidthRelative = 1;
-                        _b = tf.tidy(function () {
-                            var imgTensor = getImageTensor_1.getImageTensor(netInput);
-                            var _a = imgTensor.shape.slice(1), height = _a[0], width = _a[1];
-                            imageDimensions = { width: width, height: height };
-                            imgTensor = padToSquare_1.padToSquare(imgTensor);
-                            paddedHeightRelative = imgTensor.shape[1] / height;
-                            paddedWidthRelative = imgTensor.shape[2] / width;
-                            return _this.forwardTensor(imgTensor);
-                        }), _boxes = _b.boxes, _scores = _b.scores;
+            var netInput, _a, _boxes, _scores, boxes, scores, i, scoresData, _b, _c, iouThreshold, indices, paddedHeightRelative, paddedWidthRelative, results;
+            return tslib_1.__generator(this, function (_d) {
+                switch (_d.label) {
+                    case 0: return [4 /*yield*/, toNetInput_1.toNetInput(input, true)];
+                    case 1:
+                        netInput = _d.sent();
+                        _a = this.forwardInput(netInput), _boxes = _a.boxes, _scores = _a.scores;
                         boxes = _boxes[0];
                         scores = _scores[0];
                         for (i = 1; i < _boxes.length; i++) {
                             _boxes[i].dispose();
                             _scores[i].dispose();
                         }
-                        _d = (_c = Array).from;
+                        _c = (_b = Array).from;
                         return [4 /*yield*/, scores.data()];
-                    case 4:
-                        scoresData = _d.apply(_c, [_e.sent()]);
+                    case 2:
+                        scoresData = _c.apply(_b, [_d.sent()]);
                         iouThreshold = 0.5;
                         indices = nonMaxSuppression_1.nonMaxSuppression(boxes, scoresData, maxResults, iouThreshold, minConfidence);
+                        paddedHeightRelative = (netInput.getPaddings(0).y + netInput.getInputHeight(0)) / netInput.getInputHeight(0);
+                        paddedWidthRelative = (netInput.getPaddings(0).x + netInput.getInputWidth(0)) / netInput.getInputWidth(0);
                         results = indices
                             .map(function (idx) {
                             var _a = [
@@ -127,7 +100,10 @@ var FaceDetectionNet = /** @class */ (function () {
                                 Math.max(0, boxes.get(idx, 1)),
                                 Math.min(1.0, boxes.get(idx, 3))
                             ].map(function (val) { return val * paddedWidthRelative; }), left = _b[0], right = _b[1];
-                            return new FaceDetection_1.FaceDetection(scoresData[idx], new Rect_1.Rect(left, top, right - left, bottom - top), imageDimensions);
+                            return new FaceDetection_1.FaceDetection(scoresData[idx], new Rect_1.Rect(left, top, right - left, bottom - top), {
+                                height: netInput.getInputHeight(0),
+                                width: netInput.getInputWidth(0)
+                            });
                         });
                         boxes.dispose();
                         scores.dispose();

build/faceDetectionNet/FaceDetectionNet.js.map

-{"version":3,"file":"FaceDetectionNet.js","sourceRoot":"","sources":["../../src/faceDetectionNet/FaceDetectionNet.ts"],"names":[],"mappings":";;;AAAA,0CAA4C;AAE5C,4DAA2D;AAE3D,8CAA6C;AAC7C,gCAA+B;AAC/B,4CAA2C;AAE3C,iDAAgD;AAChD,iDAAgD;AAChD,6DAA4D;AAC5D,6CAA4C;AAC5C,yDAAwD;AACxD,6CAA4C;AAC5C,qDAAoD;AACpD,6CAA4C;AAG5C;IAAA;IA6HA,CAAC;IAzHc,+BAAI,GAAjB,UAAkB,YAAoC;;;;;;wBACpD,IAAI,YAAY,YAAY,YAAY,EAAE;4BACxC,IAAI,CAAC,cAAc,CAAC,YAAY,CAAC,CAAA;4BACjC,sBAAM;yBACP;wBAED,IAAI,YAAY,IAAI,OAAO,YAAY,KAAK,QAAQ,EAAE;4BACpD,MAAM,IAAI,KAAK,CAAC,wEAAwE,CAAC,CAAA;yBAC1F;wBACD,KAAA,IAAI,CAAA;wBAAW,qBAAM,yCAAmB,CAAC,YAAY,CAAC,EAAA;;wBAAtD,GAAK,OAAO,GAAG,SAAuC,CAAA;;;;;KACvD;IAEM,yCAAc,GAArB,UAAsB,OAAqB;QACzC,IAAI,CAAC,OAAO,GAAG,6BAAa,CAAC,OAAO,CAAC,CAAA;IACvC,CAAC;IAEO,wCAAa,GAArB,UAAsB,SAAsB;QAA5C,iBAiBC;QAhBC,IAAI,CAAC,IAAI,CAAC,OAAO,EAAE;YACjB,MAAM,IAAI,KAAK,CAAC,gDAAgD,CAAC,CAAA;SAClE;QAED,OAAO,EAAE,CAAC,IAAI,CAAC;YAEb,IAAM,OAAO,GAAG,yBAAW,CAAC,SAAS,CAAgB,CAAA;YACrD,IAAM,QAAQ,GAAG,yBAAW,CAAC,OAAO,EAAE,KAAI,CAAC,OAAO,CAAC,kBAAkB,CAAC,CAAA;YAEhE,IAAA,4GAGkF,EAFtF,kCAAc,EACd,sCAAgB,CACsE;YAExF,OAAO,yBAAW,CAAC,cAAc,EAAE,gBAAgB,EAAE,KAAI,CAAC,OAAO,CAAC,mBAAmB,CAAC,CAAA;QACxF,CAAC,CAAC,CAAA;IACJ,CAAC;IAEY,kCAAO,GAApB,UAAqB,KAAuC;;;;;;;6BACzC,CAAA,KAAK,YAAY,EAAE,CAAC,MAAM,CAAA,EAA1B,wBAA0B;wBACvC,KAAA,KAAK,CAAA;;4BACL,qBAAM,uBAAU,CAAC,KAAK,CAAC,EAAA;;wBAAvB,KAAA,SAAuB,CAAA;;;wBAFrB,QAAQ,KAEa;wBAE3B,sBAAO,EAAE,CAAC,IAAI,CAAC;gCACb,OAAA,KAAI,CAAC,aAAa,CAAC,yBAAW,CAAC,+BAAc,CAAC,QAAQ,CAAC,CAAC,CAAC;4BAAzD,CAAyD,CAC1D,EAAA;;;;KACF;IAEY,sCAAW,GAAxB,UACE,KAAuC,EACvC,aAA2B,EAC3B,UAAwB;QADxB,8BAAA,EAAA,mBAA2B;QAC3B,2BAAA,EAAA,gBAAwB;;;;;;;6BAGP,CAAA,KAAK,YAAY,EAAE,CAAC,MAAM,CAAA,EAA1B,wBAA0B;wBACvC,KAAA,KAAK,CAAA;;4BACL,qBAAM,uBAAU,CAAC,KAAK,CAAC,EAAA;;wBAAvB,KAAA,SAAuB,CAAA;;;wBAFrB,QAAQ,KAEa;wBAEvB,oBAAoB,GAAG,CAAC,EAAE,mBAAmB,GAAG,CAAC,CAAA;wBAG/C,KAGF,EAAE,CAAC,IAAI,CAAC;4BAEV,IAAI,SAAS,GAAG,+BAAc,CAAC,QAAQ,CAAC,CAAA;4BAClC,IAAA,6BAA0C,EAAzC,cAAM,EAAE,aAAK,CAA4B;4BAChD,eAAe,GAAG,EAAE,KAAK,OAAA,EAAE,MAAM,QAAA,EAAE,CAAA;4BAEnC,SAAS,GAAG,yBAAW,CAAC,SAAS,CAAC,CAAA;4BAClC,oBAAoB,GAAG,SAAS,CAAC,KAAK,CAAC,CAAC,CAAC,GAAG,MAAM,CAAA;4BAClD,mBAAmB,GAAG,SAAS,CAAC,KAAK,CAAC,CAAC,CAAC,GAAG,KAAK,CAAA;4BAEhD,OAAO,KAAI,CAAC,aAAa,CAAC,SAAS,CAAC,CAAA;wBACtC,CAAC,CAAC,EAbO,MAAM,WAAA,EACL,OAAO,YAAA,CAYf;wBAGI,KAAK,GAAG,MAAM,CAAC,CAAC,CAAC,CAAA;wBACjB,MAAM,GAAG,OAAO,CAAC,CAAC,CAAC,CAAA;wBACzB,KAAS,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE;4BACtC,MAAM,CAAC,CAAC,CAAC,CAAC,OAAO,EAAE,CAAA;4BACnB,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,EAAE,CAAA;yBACrB;wBAGkB,KAAA,CAAA,KAAA,KAAK,CAAA,CAAC,IAAI,CAAA;wBAAC,qBAAM,MAAM,CAAC,IAAI,EAAE,EAAA;;wBAA3C,UAAU,GAAG,cAAW,SAAmB,EAAC;wBAE5C,YAAY,GAAG,GAAG,CAAA;wBAClB,OAAO,GAAG,qCAAiB,CAC/B,KAAK,EACL,UAAU,EACV,UAAU,EACV,YAAY,EACZ,aAAa,CACd,CAAA;wBAEK,OAAO,GAAG,OAAO;6BACpB,GAAG,CAAC,UAAA,GAAG;4BACA,IAAA;;;wFAGkC,EAHjC,WAAG,EAAE,cAAM,CAGsB;4BAClC,IAAA;;;uFAGiC,EAHhC,YAAI,EAAE,aAAK,CAGqB;4BACvC,OAAO,IAAI,6BAAa,CACtB,UAAU,CAAC,GAAG,CAAC,EACf,IAAI,WAAI,CACN,IAAI,EACJ,GAAG,EACH,KAAK,GAAG,IAAI,EACZ,MAAM,GAAG,GAAG,CACb,EACD,eAA6B,CAC9B,CAAA;wBACH,CAAC,CAAC,CAAA;wBAEJ,KAAK,CAAC,OAAO,EAAE,CAAA;wBACf,MAAM,CAAC,OAAO,EAAE,CAAA;wBAEhB,sBAAO,OAAO,EAAA;;;;KACf;IACH,uBAAC;AAAD,CAAC,AA7HD,IA6HC;AA7HY,4CAAgB"}
\ No newline at end of file
+{"version":3,"file":"FaceDetectionNet.js","sourceRoot":"","sources":["../../src/faceDetectionNet/FaceDetectionNet.ts"],"names":[],"mappings":";;;AAAA,0CAA4C;AAG5C,gCAA+B;AAC/B,4CAA2C;AAE3C,iDAAgD;AAChD,iDAAgD;AAChD,6DAA4D;AAC5D,6CAA4C;AAC5C,yDAAwD;AACxD,6CAA4C;AAC5C,qDAAoD;AAGpD;IAAA;IAgHA,CAAC;IA5Gc,+BAAI,GAAjB,UAAkB,YAAoC;;;;;;wBACpD,IAAI,YAAY,YAAY,YAAY,EAAE;4BACxC,IAAI,CAAC,cAAc,CAAC,YAAY,CAAC,CAAA;4BACjC,sBAAM;yBACP;wBAED,IAAI,YAAY,IAAI,OAAO,YAAY,KAAK,QAAQ,EAAE;4BACpD,MAAM,IAAI,KAAK,CAAC,wEAAwE,CAAC,CAAA;yBAC1F;wBACD,KAAA,IAAI,CAAA;wBAAW,qBAAM,yCAAmB,CAAC,YAAY,CAAC,EAAA;;wBAAtD,GAAK,OAAO,GAAG,SAAuC,CAAA;;;;;KACvD;IAEM,yCAAc,GAArB,UAAsB,OAAqB;QACzC,IAAI,CAAC,OAAO,GAAG,6BAAa,CAAC,OAAO,CAAC,CAAA;IACvC,CAAC;IAEM,uCAAY,GAAnB,UAAoB,KAAe;QAAnC,iBAkBC;QAjBC,IAAI,CAAC,IAAI,CAAC,OAAO,EAAE;YACjB,MAAM,IAAI,KAAK,CAAC,gDAAgD,CAAC,CAAA;SAClE;QAED,OAAO,EAAE,CAAC,IAAI,CAAC;YACb,IAAM,WAAW,GAAG,KAAK,CAAC,aAAa,CAAC,GAAG,EAAE,KAAK,CAAC,CAAA;YAEnD,IAAM,CAAC,GAAG,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,WAAW,EAAE,EAAE,CAAC,MAAM,CAAC,oBAAoB,CAAC,CAAC,EAAE,EAAE,CAAC,MAAM,CAAC,CAAC,CAAC,CAAgB,CAAA;YACnG,IAAM,QAAQ,GAAG,yBAAW,CAAC,CAAC,EAAE,KAAI,CAAC,OAAO,CAAC,kBAAkB,CAAC,CAAA;YAE1D,IAAA,4GAGkF,EAFtF,kCAAc,EACd,sCAAgB,CACsE;YAExF,OAAO,yBAAW,CAAC,cAAc,EAAE,gBAAgB,EAAE,KAAI,CAAC,OAAO,CAAC,mBAAmB,CAAC,CAAA;QACxF,CAAC,CAAC,CAAA;IACJ,CAAC;IAEY,kCAAO,GAApB,UAAqB,KAAgB;;;;;;wBAC5B,KAAA,IAAI,CAAC,YAAY,CAAA;wBAAC,qBAAM,uBAAU,CAAC,KAAK,EAAE,IAAI,CAAC,EAAA;4BAAtD,sBAAO,SAAA,IAAI,GAAc,SAA6B,EAAC,EAAA;;;;KACxD;IAEY,sCAAW,GAAxB,UACE,KAAgB,EAChB,aAA2B,EAC3B,UAAwB;QADxB,8BAAA,EAAA,mBAA2B;QAC3B,2BAAA,EAAA,gBAAwB;;;;;4BAGP,qBAAM,uBAAU,CAAC,KAAK,EAAE,IAAI,CAAC,EAAA;;wBAAxC,QAAQ,GAAG,SAA6B;wBAExC,KAGF,IAAI,CAAC,YAAY,CAAC,QAAQ,CAAC,EAFtB,MAAM,WAAA,EACL,OAAO,YAAA,CACc;wBAIzB,KAAK,GAAG,MAAM,CAAC,CAAC,CAAC,CAAA;wBACjB,MAAM,GAAG,OAAO,CAAC,CAAC,CAAC,CAAA;wBACzB,KAAS,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE;4BACtC,MAAM,CAAC,CAAC,CAAC,CAAC,OAAO,EAAE,CAAA;4BACnB,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,EAAE,CAAA;yBACrB;wBAGkB,KAAA,CAAA,KAAA,KAAK,CAAA,CAAC,IAAI,CAAA;wBAAC,qBAAM,MAAM,CAAC,IAAI,EAAE,EAAA;;wBAA3C,UAAU,GAAG,cAAW,SAAmB,EAAC;wBAE5C,YAAY,GAAG,GAAG,CAAA;wBAClB,OAAO,GAAG,qCAAiB,CAC/B,KAAK,EACL,UAAU,EACV,UAAU,EACV,YAAY,EACZ,aAAa,CACd,CAAA;wBAGK,oBAAoB,GAAG,CAAC,QAAQ,CAAC,WAAW,CAAC,CAAC,CAAC,CAAC,CAAC,GAAG,QAAQ,CAAC,cAAc,CAAC,CAAC,CAAC,CAAC,GAAG,QAAQ,CAAC,cAAc,CAAC,CAAC,CAAC,CAAA;wBAC5G,mBAAmB,GAAG,CAAC,QAAQ,CAAC,WAAW,CAAC,CAAC,CAAC,CAAC,CAAC,GAAG,QAAQ,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,GAAG,QAAQ,CAAC,aAAa,CAAC,CAAC,CAAC,CAAA;wBAEzG,OAAO,GAAG,OAAO;6BACpB,GAAG,CAAC,UAAA,GAAG;4BACA,IAAA;;;wFAGkC,EAHjC,WAAG,EAAE,cAAM,CAGsB;4BAClC,IAAA;;;uFAGiC,EAHhC,YAAI,EAAE,aAAK,CAGqB;4BACvC,OAAO,IAAI,6BAAa,CACtB,UAAU,CAAC,GAAG,CAAC,EACf,IAAI,WAAI,CACN,IAAI,EACJ,GAAG,EACH,KAAK,GAAG,IAAI,EACZ,MAAM,GAAG,GAAG,CACb,EACD;gCACE,MAAM,EAAE,QAAQ,CAAC,cAAc,CAAC,CAAC,CAAC;gCAClC,KAAK,EAAE,QAAQ,CAAC,aAAa,CAAC,CAAC,CAAC;6BACjC,CACF,CAAA;wBACH,CAAC,CAAC,CAAA;wBAEJ,KAAK,CAAC,OAAO,EAAE,CAAA;wBACf,MAAM,CAAC,OAAO,EAAE,CAAA;wBAEhB,sBAAO,OAAO,EAAA;;;;KACf;IACH,uBAAC;AAAD,CAAC,AAhHD,IAgHC;AAhHY,4CAAgB"}
\ No newline at end of file

build/faceDetectionNet/resizeLayer.d.ts

-import * as tf from '@tensorflow/tfjs-core';
-export declare function resizeLayer(x: tf.Tensor4D): tf.Tensor<tf.Rank>;

build/faceDetectionNet/resizeLayer.js

-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-var tf = require("@tensorflow/tfjs-core");
-var resizedImageSize = [512, 512];
-var weight = tf.scalar(0.007843137718737125);
-var bias = tf.scalar(1);
-function resizeLayer(x) {
-    return tf.tidy(function () {
-        var resized = tf.image.resizeBilinear(x, resizedImageSize, false);
-        return tf.sub(tf.mul(resized, weight), bias);
-    });
-}
-exports.resizeLayer = resizeLayer;
-//# sourceMappingURL=resizeLayer.js.map
\ No newline at end of file

build/faceDetectionNet/resizeLayer.js.map

-{"version":3,"file":"resizeLayer.js","sourceRoot":"","sources":["../../src/faceDetectionNet/resizeLayer.ts"],"names":[],"mappings":";;AAAA,0CAA4C;AAE5C,IAAM,gBAAgB,GAAG,CAAC,GAAG,EAAE,GAAG,CAAqB,CAAA;AACvD,IAAM,MAAM,GAAG,EAAE,CAAC,MAAM,CAAC,oBAAoB,CAAC,CAAA;AAC9C,IAAM,IAAI,GAAG,EAAE,CAAC,MAAM,CAAC,CAAC,CAAC,CAAA;AAEzB,qBAA4B,CAAc;IACxC,OAAO,EAAE,CAAC,IAAI,CAAC;QAEb,IAAM,OAAO,GAAG,EAAE,CAAC,KAAK,CAAC,cAAc,CAAC,CAAC,EAAE,gBAAgB,EAAE,KAAK,CAAC,CAAA;QACnE,OAAO,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,OAAO,EAAE,MAAM,CAAC,EAAE,IAAI,CAAC,CAAA;IAE9C,CAAC,CAAC,CAAA;AACJ,CAAC;AAPD,kCAOC"}
\ No newline at end of file

build/faceLandmarkNet/FaceLandmarkNet.d.ts

@@ -6,7 +6,7 @@ export declare class FaceLandmarkNet {
     private _params;
     load(weightsOrUrl: Float32Array | string | undefined): Promise<void>;
     extractWeights(weights: Float32Array): void;
-    forwardTensor(imgTensor: tf.Tensor4D): tf.Tensor2D;
-    forward(input: tf.Tensor | NetInput | TNetInput): Promise<tf.Tensor2D>;
-    detectLandmarks(input: tf.Tensor | NetInput | TNetInput): Promise<FaceLandmarks>;
+    forwardInput(input: NetInput): tf.Tensor2D;
+    forward(input: TNetInput): Promise<tf.Tensor2D>;
+    detectLandmarks(input: TNetInput): Promise<FaceLandmarks | FaceLandmarks[]>;
 }

build/faceLandmarkNet/FaceLandmarkNet.js

@@ -3,8 +3,6 @@ Object.defineProperty(exports, "__esModule", { value: true });
 var tslib_1 = require("tslib");
 var tf = require("@tensorflow/tfjs-core");
 var convLayer_1 = require("../commons/convLayer");
-var getImageTensor_1 = require("../commons/getImageTensor");
-var padToSquare_1 = require("../padToSquare");
 var Point_1 = require("../Point");
 var toNetInput_1 = require("../toNetInput");
 var utils_1 = require("../utils");
@@ -47,20 +45,14 @@ var FaceLandmarkNet = /** @class */ (function () {
     FaceLandmarkNet.prototype.extractWeights = function (weights) {
         this._params = extractParams_1.extractParams(weights);
     };
-    FaceLandmarkNet.prototype.forwardTensor = function (imgTensor) {
+    FaceLandmarkNet.prototype.forwardInput = function (input) {
         var params = this._params;
         if (!params) {
             throw new Error('FaceLandmarkNet - load model before inference');
         }
         return tf.tidy(function () {
-            var _a = imgTensor.shape.slice(), batchSize = _a[0], height = _a[1], width = _a[2];
-            var x = padToSquare_1.padToSquare(imgTensor, true);
-            var _b = x.shape.slice(1), heightAfterPadding = _b[0], widthAfterPadding = _b[1];
-            // work with 128 x 128 sized face images
-            if (heightAfterPadding !== 128 || widthAfterPadding !== 128) {
-                x = tf.image.resizeBilinear(x, [128, 128]);
-            }
-            var out = conv(x, params.conv0_params);
+            var batchTensor = input.toBatchTensor(128, true);
+            var out = conv(batchTensor, params.conv0_params);
             out = maxPool(out);
             out = conv(out, params.conv1_params);
             out = conv(out, params.conv2_params);
@@ -78,37 +70,34 @@ var FaceLandmarkNet = /** @class */ (function () {
                 return tf.stack([
                     tf.fill([68], fillX),
                     tf.fill([68], fillY)
-                ], 1).as2D(batchSize, 136);
+                ], 1).as2D(1, 136).as1D();
             };
             /* shift coordinates back, to undo centered padding
-              ((x * widthAfterPadding) - shiftX) / width
-              ((y * heightAfterPadding) - shiftY) / height
+              x = ((x * widthAfterPadding) - shiftX) / width
+              y = ((y * heightAfterPadding) - shiftY) / height
             */
-            var shiftX = Math.floor(Math.abs(widthAfterPadding - width) / 2);
-            var shiftY = Math.floor(Math.abs(heightAfterPadding - height) / 2);
-            var landmarkTensor = fc1
-                .mul(createInterleavedTensor(widthAfterPadding, heightAfterPadding))
-                .sub(createInterleavedTensor(shiftX, shiftY))
-                .div(createInterleavedTensor(width, height));
-            return landmarkTensor;
+            var landmarkTensors = fc1
+                .mul(tf.stack(Array.from(Array(input.batchSize), function (_, batchIdx) {
+                return createInterleavedTensor(input.getPaddings(batchIdx).x + input.getInputWidth(batchIdx), input.getPaddings(batchIdx).y + input.getInputHeight(batchIdx));
+            })))
+                .sub(tf.stack(Array.from(Array(input.batchSize), function (_, batchIdx) {
+                return createInterleavedTensor(Math.floor(input.getPaddings(batchIdx).x / 2), Math.floor(input.getPaddings(batchIdx).y / 2));
+            })))
+                .div(tf.stack(Array.from(Array(input.batchSize), function (_, batchIdx) {
+                return createInterleavedTensor(input.getInputWidth(batchIdx), input.getInputHeight(batchIdx));
+            })));
+            return landmarkTensors;
         });
     };
     FaceLandmarkNet.prototype.forward = function (input) {
         return tslib_1.__awaiter(this, void 0, void 0, function () {
-            var netInput, _a;
+            var _a;
             return tslib_1.__generator(this, function (_b) {
                 switch (_b.label) {
                     case 0:
-                        if (!(input instanceof tf.Tensor)) return [3 /*break*/, 1];
-                        _a = input;
-                        return [3 /*break*/, 3];
-                    case 1: return [4 /*yield*/, toNetInput_1.toNetInput(input)];
-                    case 2:
-                        _a = _b.sent();
-                        _b.label = 3;
-                    case 3:
-                        netInput = _a;
-                        return [2 /*return*/, this.forwardTensor(getImageTensor_1.getImageTensor(netInput))];
+                        _a = this.forwardInput;
+                        return [4 /*yield*/, toNetInput_1.toNetInput(input, true)];
+                    case 1: return [2 /*return*/, _a.apply(this, [_b.sent()])];
                 }
             });
         });
@@ -116,33 +105,37 @@ var FaceLandmarkNet = /** @class */ (function () {
     FaceLandmarkNet.prototype.detectLandmarks = function (input) {
         return tslib_1.__awaiter(this, void 0, void 0, function () {
             var _this = this;
-            var netInput, _a, imageDimensions, outTensor, faceLandmarksArray, _b, _c, xCoords, yCoords;
-            return tslib_1.__generator(this, function (_d) {
-                switch (_d.label) {
-                    case 0:
-                        if (!(input instanceof tf.Tensor)) return [3 /*break*/, 1];
-                        _a = input;
-                        return [3 /*break*/, 3];
-                    case 1: return [4 /*yield*/, toNetInput_1.toNetInput(input)];
+            var netInput, landmarkTensors, landmarksForBatch;
+            return tslib_1.__generator(this, function (_a) {
+                switch (_a.label) {
+                    case 0: return [4 /*yield*/, toNetInput_1.toNetInput(input, true)];
+                    case 1:
+                        netInput = _a.sent();
+                        landmarkTensors = tf.tidy(function () { return tf.unstack(_this.forwardInput(netInput)); });
+                        return [4 /*yield*/, Promise.all(landmarkTensors.map(function (landmarkTensor, batchIdx) { return tslib_1.__awaiter(_this, void 0, void 0, function () {
+                                var landmarksArray, _a, _b, xCoords, yCoords;
+                                return tslib_1.__generator(this, function (_c) {
+                                    switch (_c.label) {
+                                        case 0:
+                                            _b = (_a = Array).from;
+                                            return [4 /*yield*/, landmarkTensor.data()];
+                                        case 1:
+                                            landmarksArray = _b.apply(_a, [_c.sent()]);
+                                            xCoords = landmarksArray.filter(function (_, i) { return utils_1.isEven(i); });
+                                            yCoords = landmarksArray.filter(function (_, i) { return !utils_1.isEven(i); });
+                                            return [2 /*return*/, new FaceLandmarks_1.FaceLandmarks(Array(68).fill(0).map(function (_, i) { return new Point_1.Point(xCoords[i], yCoords[i]); }), {
+                                                    height: netInput.getInputHeight(batchIdx),
+                                                    width: netInput.getInputWidth(batchIdx),
+                                                })];
+                                    }
+                                });
+                            }); }))];
                     case 2:
-                        _a = _d.sent();
-                        _d.label = 3;
-                    case 3:
-                        netInput = _a;
-                        outTensor = tf.tidy(function () {
-                            var imgTensor = getImageTensor_1.getImageTensor(netInput);
-                            var _a = imgTensor.shape.slice(1), height = _a[0], width = _a[1];
-                            imageDimensions = { width: width, height: height };
-                            return _this.forwardTensor(imgTensor);
-                        });
-                        _c = (_b = Array).from;
-                        return [4 /*yield*/, outTensor.data()];
-                    case 4:
-                        faceLandmarksArray = _c.apply(_b, [_d.sent()]);
-                        outTensor.dispose();
-                        xCoords = faceLandmarksArray.filter(function (_, i) { return utils_1.isEven(i); });
-                        yCoords = faceLandmarksArray.filter(function (_, i) { return !utils_1.isEven(i); });
-                        return [2 /*return*/, new FaceLandmarks_1.FaceLandmarks(Array(68).fill(0).map(function (_, i) { return new Point_1.Point(xCoords[i], yCoords[i]); }), imageDimensions)];
+                        landmarksForBatch = _a.sent();
+                        landmarkTensors.forEach(function (t) { return t.dispose(); });
+                        return [2 /*return*/, netInput.isBatchInput
+                                ? landmarksForBatch
+                                : landmarksForBatch[0]];
                 }
             });
         });

build/faceLandmarkNet/FaceLandmarkNet.js.map

-{"version":3,"file":"FaceLandmarkNet.js","sourceRoot":"","sources":["../../src/faceLandmarkNet/FaceLandmarkNet.ts"],"names":[],"mappings":";;;AAAA,0CAA4C;AAE5C,kDAAiD;AACjD,4DAA2D;AAG3D,8CAA6C;AAC7C,kCAAiC;AACjC,4CAA2C;AAE3C,kCAAkC;AAClC,iDAAgD;AAChD,iDAAgD;AAChD,6DAA4D;AAC5D,6DAA4D;AAG5D,cAAc,CAAc,EAAE,MAAkB;IAC9C,OAAO,qBAAS,CAAC,CAAC,EAAE,MAAM,EAAE,OAAO,EAAE,IAAI,CAAC,CAAA;AAC5C,CAAC;AAED,iBAAiB,CAAc,EAAE,OAAkC;IAAlC,wBAAA,EAAA,WAA6B,CAAC,EAAE,CAAC,CAAC;IACjE,OAAO,EAAE,CAAC,OAAO,CAAC,CAAC,EAAE,CAAC,CAAC,EAAE,CAAC,CAAC,EAAE,OAAO,EAAE,OAAO,CAAC,CAAA;AAChD,CAAC;AAED;IAAA;IA+GA,CAAC;IA3Gc,8BAAI,GAAjB,UAAkB,YAA+C;;;;;;wBAC/D,IAAI,YAAY,YAAY,YAAY,EAAE;4BACxC,IAAI,CAAC,cAAc,CAAC,YAAY,CAAC,CAAA;4BACjC,sBAAM;yBACP;wBAED,IAAI,YAAY,IAAI,OAAO,YAAY,KAAK,QAAQ,EAAE;4BACpD,MAAM,IAAI,KAAK,CAAC,uEAAuE,CAAC,CAAA;yBACzF;wBACD,KAAA,IAAI,CAAA;wBAAW,qBAAM,yCAAmB,CAAC,YAAY,CAAC,EAAA;;wBAAtD,GAAK,OAAO,GAAG,SAAuC,CAAA;;;;;KACvD;IAEM,wCAAc,GAArB,UAAsB,OAAqB;QACzC,IAAI,CAAC,OAAO,GAAG,6BAAa,CAAC,OAAO,CAAC,CAAA;IACvC,CAAC;IAEM,uCAAa,GAApB,UAAqB,SAAsB;QACzC,IAAM,MAAM,GAAG,IAAI,CAAC,OAAO,CAAA;QAE3B,IAAI,CAAC,MAAM,EAAE;YACX,MAAM,IAAI,KAAK,CAAC,+CAA+C,CAAC,CAAA;SACjE;QAED,OAAO,EAAE,CAAC,IAAI,CAAC;YACP,IAAA,4BAAoD,EAAnD,iBAAS,EAAE,cAAM,EAAE,aAAK,CAA2B;YAE1D,IAAI,CAAC,GAAG,yBAAW,CAAC,SAAS,EAAE,IAAI,CAAC,CAAA;YAC9B,IAAA,qBAA0D,EAAzD,0BAAkB,EAAE,yBAAiB,CAAoB;YAEhE,wCAAwC;YACxC,IAAI,kBAAkB,KAAK,GAAG,IAAI,iBAAiB,KAAK,GAAG,EAAE;gBAC3D,CAAC,GAAG,EAAE,CAAC,KAAK,CAAC,cAAc,CAAC,CAAC,EAAE,CAAC,GAAG,EAAE,GAAG,CAAC,CAAC,CAAA;aAC3C;YAED,IAAI,GAAG,GAAG,IAAI,CAAC,CAAC,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACtC,GAAG,GAAG,OAAO,CAAC,GAAG,CAAC,CAAA;YAClB,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,OAAO,CAAC,GAAG,CAAC,CAAA;YAClB,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,OAAO,CAAC,GAAG,CAAC,CAAA;YAClB,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,OAAO,CAAC,GAAG,EAAE,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAA;YAC1B,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,IAAM,GAAG,GAAG,EAAE,CAAC,IAAI,CAAC,yCAAmB,CAAC,GAAG,CAAC,IAAI,CAAC,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,MAAM,CAAC,UAAU,CAAC,CAAC,CAAA;YACvF,IAAM,GAAG,GAAG,yCAAmB,CAAC,GAAG,EAAE,MAAM,CAAC,UAAU,CAAC,CAAA;YAGvD,IAAM,uBAAuB,GAAG,UAAC,KAAa,EAAE,KAAa;gBAC7D,OAAA,EAAE,CAAC,KAAK,CAAC;oBACP,EAAE,CAAC,IAAI,CAAC,CAAC,EAAE,CAAC,EAAE,KAAK,CAAC;oBACpB,EAAE,CAAC,IAAI,CAAC,CAAC,EAAE,CAAC,EAAE,KAAK,CAAC;iBACrB,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,SAAS,EAAE,GAAG,CAAC;YAH1B,CAG0B,CAAA;YAG1B;;;cAGE;YACF,IAAM,MAAM,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,iBAAiB,GAAG,KAAK,CAAC,GAAG,CAAC,CAAC,CAAA;YAClE,IAAM,MAAM,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,kBAAkB,GAAG,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YACpE,IAAM,cAAc,GAAG,GAAG;iBACvB,GAAG,CAAC,uBAAuB,CAAC,iBAAiB,EAAE,kBAAkB,CAAC,CAAC;iBACnE,GAAG,CAAC,uBAAuB,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;iBAC5C,GAAG,CAAC,uBAAuB,CAAC,KAAK,EAAE,MAAM,CAAC,CAAC,CAAA;YAE9C,OAAO,cAA6B,CAAA;QACtC,CAAC,CAAC,CAAA;IACJ,CAAC;IAEY,iCAAO,GAApB,UAAqB,KAAuC;;;;;;6BACzC,CAAA,KAAK,YAAY,EAAE,CAAC,MAAM,CAAA,EAA1B,wBAA0B;wBACvC,KAAA,KAAK,CAAA;;4BACL,qBAAM,uBAAU,CAAC,KAAK,CAAC,EAAA;;wBAAvB,KAAA,SAAuB,CAAA;;;wBAFrB,QAAQ,KAEa;wBAE3B,sBAAO,IAAI,CAAC,aAAa,CAAC,+BAAc,CAAC,QAAQ,CAAC,CAAC,EAAA;;;;KACpD;IAEY,yCAAe,GAA5B,UAA6B,KAAuC;;;;;;;6BACjD,CAAA,KAAK,YAAY,EAAE,CAAC,MAAM,CAAA,EAA1B,wBAA0B;wBACvC,KAAA,KAAK,CAAA;;4BACL,qBAAM,uBAAU,CAAC,KAAK,CAAC,EAAA;;wBAAvB,KAAA,SAAuB,CAAA;;;wBAFrB,QAAQ,KAEa;wBAIrB,SAAS,GAAG,EAAE,CAAC,IAAI,CAAC;4BACxB,IAAM,SAAS,GAAG,+BAAc,CAAC,QAAQ,CAAC,CAAA;4BAEpC,IAAA,6BAA0C,EAAzC,cAAM,EAAE,aAAK,CAA4B;4BAChD,eAAe,GAAG,EAAE,KAAK,OAAA,EAAE,MAAM,QAAA,EAAE,CAAA;4BAEnC,OAAO,KAAI,CAAC,aAAa,CAAC,SAAS,CAAC,CAAA;wBACtC,CAAC,CAAC,CAAA;wBAEyB,KAAA,CAAA,KAAA,KAAK,CAAA,CAAC,IAAI,CAAA;wBAAC,qBAAM,SAAS,CAAC,IAAI,EAAE,EAAA;;wBAAtD,kBAAkB,GAAG,cAAW,SAAsB,EAAC;wBAC7D,SAAS,CAAC,OAAO,EAAE,CAAA;wBAEb,OAAO,GAAG,kBAAkB,CAAC,MAAM,CAAC,UAAC,CAAC,EAAE,CAAC,IAAK,OAAA,cAAM,CAAC,CAAC,CAAC,EAAT,CAAS,CAAC,CAAA;wBACxD,OAAO,GAAG,kBAAkB,CAAC,MAAM,CAAC,UAAC,CAAC,EAAE,CAAC,IAAK,OAAA,CAAC,cAAM,CAAC,CAAC,CAAC,EAAV,CAAU,CAAC,CAAA;wBAE/D,sBAAO,IAAI,6BAAa,CACtB,KAAK,CAAC,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,UAAC,CAAC,EAAE,CAAC,IAAK,OAAA,IAAI,aAAK,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,OAAO,CAAC,CAAC,CAAC,CAAC,EAAjC,CAAiC,CAAC,EAClE,eAA6B,CAC9B,EAAA;;;;KACF;IACH,sBAAC;AAAD,CAAC,AA/GD,IA+GC;AA/GY,0CAAe"}
\ No newline at end of file
+{"version":3,"file":"FaceLandmarkNet.js","sourceRoot":"","sources":["../../src/faceLandmarkNet/FaceLandmarkNet.ts"],"names":[],"mappings":";;;AAAA,0CAA4C;AAE5C,kDAAiD;AAGjD,kCAAiC;AACjC,4CAA2C;AAE3C,kCAAkC;AAClC,iDAAgD;AAChD,iDAAgD;AAChD,6DAA4D;AAC5D,6DAA4D;AAG5D,cAAc,CAAc,EAAE,MAAkB;IAC9C,OAAO,qBAAS,CAAC,CAAC,EAAE,MAAM,EAAE,OAAO,EAAE,IAAI,CAAC,CAAA;AAC5C,CAAC;AAED,iBAAiB,CAAc,EAAE,OAAkC;IAAlC,wBAAA,EAAA,WAA6B,CAAC,EAAE,CAAC,CAAC;IACjE,OAAO,EAAE,CAAC,OAAO,CAAC,CAAC,EAAE,CAAC,CAAC,EAAE,CAAC,CAAC,EAAE,OAAO,EAAE,OAAO,CAAC,CAAA;AAChD,CAAC;AAED;IAAA;IAiHA,CAAC;IA7Gc,8BAAI,GAAjB,UAAkB,YAA+C;;;;;;wBAC/D,IAAI,YAAY,YAAY,YAAY,EAAE;4BACxC,IAAI,CAAC,cAAc,CAAC,YAAY,CAAC,CAAA;4BACjC,sBAAM;yBACP;wBAED,IAAI,YAAY,IAAI,OAAO,YAAY,KAAK,QAAQ,EAAE;4BACpD,MAAM,IAAI,KAAK,CAAC,uEAAuE,CAAC,CAAA;yBACzF;wBACD,KAAA,IAAI,CAAA;wBAAW,qBAAM,yCAAmB,CAAC,YAAY,CAAC,EAAA;;wBAAtD,GAAK,OAAO,GAAG,SAAuC,CAAA;;;;;KACvD;IAEM,wCAAc,GAArB,UAAsB,OAAqB;QACzC,IAAI,CAAC,OAAO,GAAG,6BAAa,CAAC,OAAO,CAAC,CAAA;IACvC,CAAC;IAEM,sCAAY,GAAnB,UAAoB,KAAe;QACjC,IAAM,MAAM,GAAG,IAAI,CAAC,OAAO,CAAA;QAE3B,IAAI,CAAC,MAAM,EAAE;YACX,MAAM,IAAI,KAAK,CAAC,+CAA+C,CAAC,CAAA;SACjE;QAED,OAAO,EAAE,CAAC,IAAI,CAAC;YACb,IAAM,WAAW,GAAG,KAAK,CAAC,aAAa,CAAC,GAAG,EAAE,IAAI,CAAC,CAAA;YAElD,IAAI,GAAG,GAAG,IAAI,CAAC,WAAW,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YAChD,GAAG,GAAG,OAAO,CAAC,GAAG,CAAC,CAAA;YAClB,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,OAAO,CAAC,GAAG,CAAC,CAAA;YAClB,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,OAAO,CAAC,GAAG,CAAC,CAAA;YAClB,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,GAAG,GAAG,OAAO,CAAC,GAAG,EAAE,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAA;YAC1B,GAAG,GAAG,IAAI,CAAC,GAAG,EAAE,MAAM,CAAC,YAAY,CAAC,CAAA;YACpC,IAAM,GAAG,GAAG,EAAE,CAAC,IAAI,CAAC,yCAAmB,CAAC,GAAG,CAAC,IAAI,CAAC,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,MAAM,CAAC,UAAU,CAAC,CAAC,CAAA;YACvF,IAAM,GAAG,GAAG,yCAAmB,CAAC,GAAG,EAAE,MAAM,CAAC,UAAU,CAAC,CAAA;YAEvD,IAAM,uBAAuB,GAAG,UAAC,KAAa,EAAE,KAAa;gBAC3D,OAAA,EAAE,CAAC,KAAK,CAAC;oBACP,EAAE,CAAC,IAAI,CAAC,CAAC,EAAE,CAAC,EAAE,KAAK,CAAC;oBACpB,EAAE,CAAC,IAAI,CAAC,CAAC,EAAE,CAAC,EAAE,KAAK,CAAC;iBACrB,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE;YAHzB,CAGyB,CAAA;YAE3B;;;cAGE;YAEF,IAAM,eAAe,GAAG,GAAG;iBACxB,GAAG,CAAC,EAAE,CAAC,KAAK,CAAC,KAAK,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,SAAS,CAAC,EAAE,UAAC,CAAC,EAAE,QAAQ;gBAC3D,OAAA,uBAAuB,CACrB,KAAK,CAAC,WAAW,CAAC,QAAQ,CAAC,CAAC,CAAC,GAAG,KAAK,CAAC,aAAa,CAAC,QAAQ,CAAC,EAC7D,KAAK,CAAC,WAAW,CAAC,QAAQ,CAAC,CAAC,CAAC,GAAG,KAAK,CAAC,cAAc,CAAC,QAAQ,CAAC,CAC/D;YAHD,CAGC,CACF,CAAC,CAAC;iBACF,GAAG,CAAC,EAAE,CAAC,KAAK,CAAC,KAAK,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,SAAS,CAAC,EAAE,UAAC,CAAC,EAAE,QAAQ;gBAC3D,OAAA,uBAAuB,CACrB,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,WAAW,CAAC,QAAQ,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,EAC7C,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,WAAW,CAAC,QAAQ,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAC9C;YAHD,CAGC,CACF,CAAC,CAAC;iBACF,GAAG,CAAC,EAAE,CAAC,KAAK,CAAC,KAAK,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,SAAS,CAAC,EAAE,UAAC,CAAC,EAAE,QAAQ;gBAC3D,OAAA,uBAAuB,CACrB,KAAK,CAAC,aAAa,CAAC,QAAQ,CAAC,EAC7B,KAAK,CAAC,cAAc,CAAC,QAAQ,CAAC,CAC/B;YAHD,CAGC,CACF,CAAC,CAAC,CAAA;YAEL,OAAO,eAA8B,CAAA;QACvC,CAAC,CAAC,CAAA;IACJ,CAAC;IAEY,iCAAO,GAApB,UAAqB,KAAgB;;;;;;wBAC5B,KAAA,IAAI,CAAC,YAAY,CAAA;wBAAC,qBAAM,uBAAU,CAAC,KAAK,EAAE,IAAI,CAAC,EAAA;4BAAtD,sBAAO,SAAA,IAAI,GAAc,SAA6B,EAAC,EAAA;;;;KACxD;IAEY,yCAAe,GAA5B,UAA6B,KAAgB;;;;;;4BAC1B,qBAAM,uBAAU,CAAC,KAAK,EAAE,IAAI,CAAC,EAAA;;wBAAxC,QAAQ,GAAG,SAA6B;wBAExC,eAAe,GAAG,EAAE,CAAC,IAAI,CAC7B,cAAM,OAAA,EAAE,CAAC,OAAO,CAAC,KAAI,CAAC,YAAY,CAAC,QAAQ,CAAC,CAAC,EAAvC,CAAuC,CAC9C,CAAA;wBAEyB,qBAAM,OAAO,CAAC,GAAG,CAAC,eAAe,CAAC,GAAG,CAC7D,UAAO,cAAc,EAAE,QAAQ;;;;;4CACN,KAAA,CAAA,KAAA,KAAK,CAAA,CAAC,IAAI,CAAA;4CAAC,qBAAM,cAAc,CAAC,IAAI,EAAE,EAAA;;4CAAvD,cAAc,GAAG,cAAW,SAA2B,EAAC;4CACxD,OAAO,GAAG,cAAc,CAAC,MAAM,CAAC,UAAC,CAAC,EAAE,CAAC,IAAK,OAAA,cAAM,CAAC,CAAC,CAAC,EAAT,CAAS,CAAC,CAAA;4CACpD,OAAO,GAAG,cAAc,CAAC,MAAM,CAAC,UAAC,CAAC,EAAE,CAAC,IAAK,OAAA,CAAC,cAAM,CAAC,CAAC,CAAC,EAAV,CAAU,CAAC,CAAA;4CAE3D,sBAAO,IAAI,6BAAa,CACtB,KAAK,CAAC,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,UAAC,CAAC,EAAE,CAAC,IAAK,OAAA,IAAI,aAAK,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,OAAO,CAAC,CAAC,CAAC,CAAC,EAAjC,CAAiC,CAAC,EAClE;oDACE,MAAM,EAAE,QAAQ,CAAC,cAAc,CAAC,QAAQ,CAAC;oDACzC,KAAK,EAAG,QAAQ,CAAC,aAAa,CAAC,QAAQ,CAAC;iDACzC,CACF,EAAA;;;iCACF,CACF,CAAC,EAAA;;wBAdI,iBAAiB,GAAG,SAcxB;wBAEF,eAAe,CAAC,OAAO,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,CAAC,OAAO,EAAE,EAAX,CAAW,CAAC,CAAA;wBAEzC,sBAAO,QAAQ,CAAC,YAAY;gCAC1B,CAAC,CAAC,iBAAiB;gCACnB,CAAC,CAAC,iBAAiB,CAAC,CAAC,CAAC,EAAA;;;;KACzB;IACH,sBAAC;AAAD,CAAC,AAjHD,IAiHC;AAjHY,0CAAe"}
\ No newline at end of file

build/faceRecognitionNet/FaceRecognitionNet.d.ts

@@ -5,6 +5,7 @@ export declare class FaceRecognitionNet {
     private _params;
     load(weightsOrUrl: Float32Array | string | undefined): Promise<void>;
     extractWeights(weights: Float32Array): void;
-    forward(input: tf.Tensor | NetInput | TNetInput): Promise<tf.Tensor2D>;
-    computeFaceDescriptor(input: tf.Tensor | NetInput | TNetInput): Promise<Float32Array>;
+    forwardInput(input: NetInput): Promise<tf.Tensor2D>;
+    forward(input: TNetInput): Promise<tf.Tensor2D>;
+    computeFaceDescriptor(input: TNetInput): Promise<Float32Array>;
 }

build/faceRecognitionNet/FaceRecognitionNet.js

@@ -2,8 +2,6 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 var tslib_1 = require("tslib");
 var tf = require("@tensorflow/tfjs-core");
-var getImageTensor_1 = require("../commons/getImageTensor");
-var padToSquare_1 = require("../padToSquare");
 var toNetInput_1 = require("../toNetInput");
 var convLayer_1 = require("./convLayer");
 var extractParams_1 = require("./extractParams");
@@ -38,76 +36,65 @@ var FaceRecognitionNet = /** @class */ (function () {
     FaceRecognitionNet.prototype.extractWeights = function (weights) {
         this._params = extractParams_1.extractParams(weights);
     };
-    FaceRecognitionNet.prototype.forward = function (input) {
+    FaceRecognitionNet.prototype.forwardInput = function (input) {
         return tslib_1.__awaiter(this, void 0, void 0, function () {
             var _this = this;
-            var netInput, _a;
+            return tslib_1.__generator(this, function (_a) {
+                if (!this._params) {
+                    throw new Error('FaceRecognitionNet - load model before inference');
+                }
+                return [2 /*return*/, tf.tidy(function () {
+                        var batchTensor = input.toBatchTensor(150, true);
+                        var normalized = normalize_1.normalize(batchTensor);
+                        var out = convLayer_1.convDown(normalized, _this._params.conv32_down);
+                        out = tf.maxPool(out, 3, 2, 'valid');
+                        out = residualLayer_1.residual(out, _this._params.conv32_1);
+                        out = residualLayer_1.residual(out, _this._params.conv32_2);
+                        out = residualLayer_1.residual(out, _this._params.conv32_3);
+                        out = residualLayer_1.residualDown(out, _this._params.conv64_down);
+                        out = residualLayer_1.residual(out, _this._params.conv64_1);
+                        out = residualLayer_1.residual(out, _this._params.conv64_2);
+                        out = residualLayer_1.residual(out, _this._params.conv64_3);
+                        out = residualLayer_1.residualDown(out, _this._params.conv128_down);
+                        out = residualLayer_1.residual(out, _this._params.conv128_1);
+                        out = residualLayer_1.residual(out, _this._params.conv128_2);
+                        out = residualLayer_1.residualDown(out, _this._params.conv256_down);
+                        out = residualLayer_1.residual(out, _this._params.conv256_1);
+                        out = residualLayer_1.residual(out, _this._params.conv256_2);
+                        out = residualLayer_1.residualDown(out, _this._params.conv256_down_out);
+                        var globalAvg = out.mean([1, 2]);
+                        var fullyConnected = tf.matMul(globalAvg, _this._params.fc);
+                        return fullyConnected;
+                    })];
+            });
+        });
+    };
+    FaceRecognitionNet.prototype.forward = function (input) {
+        return tslib_1.__awaiter(this, void 0, void 0, function () {
+            var _a;
             return tslib_1.__generator(this, function (_b) {
                 switch (_b.label) {
                     case 0:
-                        if (!this._params) {
-                            throw new Error('FaceRecognitionNet - load model before inference');
-                        }
-                        if (!(input instanceof tf.Tensor)) return [3 /*break*/, 1];
-                        _a = input;
-                        return [3 /*break*/, 3];
-                    case 1: return [4 /*yield*/, toNetInput_1.toNetInput(input)];
-                    case 2:
-                        _a = _b.sent();
-                        _b.label = 3;
-                    case 3:
-                        netInput = _a;
-                        return [2 /*return*/, tf.tidy(function () {
-                                var x = padToSquare_1.padToSquare(getImageTensor_1.getImageTensor(netInput), true);
-                                // work with 150 x 150 sized face images
-                                if (x.shape[1] !== 150 || x.shape[2] !== 150) {
-                                    x = tf.image.resizeBilinear(x, [150, 150]);
-                                }
-                                x = normalize_1.normalize(x);
-                                var out = convLayer_1.convDown(x, _this._params.conv32_down);
-                                out = tf.maxPool(out, 3, 2, 'valid');
-                                out = residualLayer_1.residual(out, _this._params.conv32_1);
-                                out = residualLayer_1.residual(out, _this._params.conv32_2);
-                                out = residualLayer_1.residual(out, _this._params.conv32_3);
-                                out = residualLayer_1.residualDown(out, _this._params.conv64_down);
-                                out = residualLayer_1.residual(out, _this._params.conv64_1);
-                                out = residualLayer_1.residual(out, _this._params.conv64_2);
-                                out = residualLayer_1.residual(out, _this._params.conv64_3);
-                                out = residualLayer_1.residualDown(out, _this._params.conv128_down);
-                                out = residualLayer_1.residual(out, _this._params.conv128_1);
-                                out = residualLayer_1.residual(out, _this._params.conv128_2);
-                                out = residualLayer_1.residualDown(out, _this._params.conv256_down);
-                                out = residualLayer_1.residual(out, _this._params.conv256_1);
-                                out = residualLayer_1.residual(out, _this._params.conv256_2);
-                                out = residualLayer_1.residualDown(out, _this._params.conv256_down_out);
-                                var globalAvg = out.mean([1, 2]);
-                                var fullyConnected = tf.matMul(globalAvg, _this._params.fc);
-                                return fullyConnected;
-                            })];
+                        _a = this.forwardInput;
+                        return [4 /*yield*/, toNetInput_1.toNetInput(input, true)];
+                    case 1: return [2 /*return*/, _a.apply(this, [_b.sent()])];
                 }
             });
         });
     };
     FaceRecognitionNet.prototype.computeFaceDescriptor = function (input) {
         return tslib_1.__awaiter(this, void 0, void 0, function () {
-            var netInput, _a, result, data;
+            var result, _a, data;
             return tslib_1.__generator(this, function (_b) {
                 switch (_b.label) {
                     case 0:
-                        if (!(input instanceof tf.Tensor)) return [3 /*break*/, 1];
-                        _a = input;
-                        return [3 /*break*/, 3];
-                    case 1: return [4 /*yield*/, toNetInput_1.toNetInput(input)];
+                        _a = this.forward;
+                        return [4 /*yield*/, toNetInput_1.toNetInput(input, true)];
+                    case 1: return [4 /*yield*/, _a.apply(this, [_b.sent()])];
                     case 2:
-                        _a = _b.sent();
-                        _b.label = 3;
-                    case 3:
-                        netInput = _a;
-                        return [4 /*yield*/, this.forward(netInput)];
-                    case 4:
                         result = _b.sent();
                         return [4 /*yield*/, result.data()];
-                    case 5:
+                    case 3:
                         data = _b.sent();
                         result.dispose();
                         return [2 /*return*/, data];

build/faceRecognitionNet/FaceRecognitionNet.js.map

-{"version":3,"file":"FaceRecognitionNet.js","sourceRoot":"","sources":["../../src/faceRecognitionNet/FaceRecognitionNet.ts"],"names":[],"mappings":";;;AAAA,0CAA4C;AAE5C,4DAA2D;AAE3D,8CAA6C;AAC7C,4CAA2C;AAE3C,yCAAuC;AACvC,iDAAgD;AAChD,6DAA4D;AAC5D,yCAAwC;AACxC,iDAAyD;AAGzD;IAAA;IA4EA,CAAC;IAxEc,iCAAI,GAAjB,UAAkB,YAA+C;;;;;;wBAC/D,IAAI,YAAY,YAAY,YAAY,EAAE;4BACxC,IAAI,CAAC,cAAc,CAAC,YAAY,CAAC,CAAA;4BACjC,sBAAM;yBACP;wBAED,IAAI,YAAY,IAAI,OAAO,YAAY,KAAK,QAAQ,EAAE;4BACpD,MAAM,IAAI,KAAK,CAAC,uEAAuE,CAAC,CAAA;yBACzF;wBACD,KAAA,IAAI,CAAA;wBAAW,qBAAM,yCAAmB,CAAC,YAAY,CAAC,EAAA;;wBAAtD,GAAK,OAAO,GAAG,SAAuC,CAAA;;;;;KACvD;IAEM,2CAAc,GAArB,UAAsB,OAAqB;QACzC,IAAI,CAAC,OAAO,GAAG,6BAAa,CAAC,OAAO,CAAC,CAAA;IACvC,CAAC;IAEY,oCAAO,GAApB,UAAqB,KAAuC;;;;;;;wBAC1D,IAAI,CAAC,IAAI,CAAC,OAAO,EAAE;4BACjB,MAAM,IAAI,KAAK,CAAC,kDAAkD,CAAC,CAAA;yBACpE;6BAEgB,CAAA,KAAK,YAAY,EAAE,CAAC,MAAM,CAAA,EAA1B,wBAA0B;wBACvC,KAAA,KAAK,CAAA;;4BACL,qBAAM,uBAAU,CAAC,KAAK,CAAC,EAAA;;wBAAvB,KAAA,SAAuB,CAAA;;;wBAFrB,QAAQ,KAEa;wBAE3B,sBAAO,EAAE,CAAC,IAAI,CAAC;gCAEb,IAAI,CAAC,GAAG,yBAAW,CAAC,+BAAc,CAAC,QAAQ,CAAC,EAAE,IAAI,CAAC,CAAA;gCACnD,wCAAwC;gCACxC,IAAI,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,KAAK,GAAG,IAAI,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,KAAK,GAAG,EAAE;oCAC5C,CAAC,GAAG,EAAE,CAAC,KAAK,CAAC,cAAc,CAAC,CAAC,EAAE,CAAC,GAAG,EAAE,GAAG,CAAC,CAAC,CAAA;iCAC3C;gCACD,CAAC,GAAG,qBAAS,CAAC,CAAC,CAAC,CAAA;gCAEhB,IAAI,GAAG,GAAG,oBAAQ,CAAC,CAAC,EAAE,KAAI,CAAC,OAAO,CAAC,WAAW,CAAC,CAAA;gCAC/C,GAAG,GAAG,EAAE,CAAC,OAAO,CAAC,GAAG,EAAE,CAAC,EAAE,CAAC,EAAE,OAAO,CAAC,CAAA;gCAEpC,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;gCAC1C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;gCAC1C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;gCAE1C,GAAG,GAAG,4BAAY,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,WAAW,CAAC,CAAA;gCACjD,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;gCAC1C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;gCAC1C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;gCAE1C,GAAG,GAAG,4BAAY,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,YAAY,CAAC,CAAA;gCAClD,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,SAAS,CAAC,CAAA;gCAC3C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,SAAS,CAAC,CAAA;gCAE3C,GAAG,GAAG,4BAAY,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,YAAY,CAAC,CAAA;gCAClD,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,SAAS,CAAC,CAAA;gCAC3C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,SAAS,CAAC,CAAA;gCAC3C,GAAG,GAAG,4BAAY,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,gBAAgB,CAAC,CAAA;gCAEtD,IAAM,SAAS,GAAG,GAAG,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAgB,CAAA;gCACjD,IAAM,cAAc,GAAG,EAAE,CAAC,MAAM,CAAC,SAAS,EAAE,KAAI,CAAC,OAAO,CAAC,EAAE,CAAC,CAAA;gCAE5D,OAAO,cAAc,CAAA;4BACvB,CAAC,CAAC,EAAA;;;;KACH;IAEY,kDAAqB,GAAlC,UAAmC,KAAuC;;;;;;6BACvD,CAAA,KAAK,YAAY,EAAE,CAAC,MAAM,CAAA,EAA1B,wBAA0B;wBACvC,KAAA,KAAK,CAAA;;4BACL,qBAAM,uBAAU,CAAC,KAAK,CAAC,EAAA;;wBAAvB,KAAA,SAAuB,CAAA;;;wBAFrB,QAAQ,KAEa;wBAEZ,qBAAM,IAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,EAAA;;wBAArC,MAAM,GAAG,SAA4B;wBAC9B,qBAAM,MAAM,CAAC,IAAI,EAAE,EAAA;;wBAA1B,IAAI,GAAG,SAAmB;wBAChC,MAAM,CAAC,OAAO,EAAE,CAAA;wBAChB,sBAAO,IAAoB,EAAA;;;;KAC5B;IACH,yBAAC;AAAD,CAAC,AA5ED,IA4EC;AA5EY,gDAAkB"}
\ No newline at end of file
+{"version":3,"file":"FaceRecognitionNet.js","sourceRoot":"","sources":["../../src/faceRecognitionNet/FaceRecognitionNet.ts"],"names":[],"mappings":";;;AAAA,0CAA4C;AAG5C,4CAA2C;AAE3C,yCAAuC;AACvC,iDAAgD;AAChD,6DAA4D;AAC5D,yCAAwC;AACxC,iDAAyD;AAGzD;IAAA;IAoEA,CAAC;IAhEc,iCAAI,GAAjB,UAAkB,YAA+C;;;;;;wBAC/D,IAAI,YAAY,YAAY,YAAY,EAAE;4BACxC,IAAI,CAAC,cAAc,CAAC,YAAY,CAAC,CAAA;4BACjC,sBAAM;yBACP;wBAED,IAAI,YAAY,IAAI,OAAO,YAAY,KAAK,QAAQ,EAAE;4BACpD,MAAM,IAAI,KAAK,CAAC,uEAAuE,CAAC,CAAA;yBACzF;wBACD,KAAA,IAAI,CAAA;wBAAW,qBAAM,yCAAmB,CAAC,YAAY,CAAC,EAAA;;wBAAtD,GAAK,OAAO,GAAG,SAAuC,CAAA;;;;;KACvD;IAEM,2CAAc,GAArB,UAAsB,OAAqB;QACzC,IAAI,CAAC,OAAO,GAAG,6BAAa,CAAC,OAAO,CAAC,CAAA;IACvC,CAAC;IAEY,yCAAY,GAAzB,UAA0B,KAAe;;;;gBACvC,IAAI,CAAC,IAAI,CAAC,OAAO,EAAE;oBACjB,MAAM,IAAI,KAAK,CAAC,kDAAkD,CAAC,CAAA;iBACpE;gBAGD,sBAAO,EAAE,CAAC,IAAI,CAAC;wBACb,IAAM,WAAW,GAAG,KAAK,CAAC,aAAa,CAAC,GAAG,EAAE,IAAI,CAAC,CAAA;wBAElD,IAAM,UAAU,GAAG,qBAAS,CAAC,WAAW,CAAC,CAAA;wBAEzC,IAAI,GAAG,GAAG,oBAAQ,CAAC,UAAU,EAAE,KAAI,CAAC,OAAO,CAAC,WAAW,CAAC,CAAA;wBACxD,GAAG,GAAG,EAAE,CAAC,OAAO,CAAC,GAAG,EAAE,CAAC,EAAE,CAAC,EAAE,OAAO,CAAC,CAAA;wBAEpC,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;wBAC1C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;wBAC1C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;wBAE1C,GAAG,GAAG,4BAAY,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,WAAW,CAAC,CAAA;wBACjD,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;wBAC1C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;wBAC1C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAA;wBAE1C,GAAG,GAAG,4BAAY,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,YAAY,CAAC,CAAA;wBAClD,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,SAAS,CAAC,CAAA;wBAC3C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,SAAS,CAAC,CAAA;wBAE3C,GAAG,GAAG,4BAAY,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,YAAY,CAAC,CAAA;wBAClD,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,SAAS,CAAC,CAAA;wBAC3C,GAAG,GAAG,wBAAQ,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,SAAS,CAAC,CAAA;wBAC3C,GAAG,GAAG,4BAAY,CAAC,GAAG,EAAE,KAAI,CAAC,OAAO,CAAC,gBAAgB,CAAC,CAAA;wBAEtD,IAAM,SAAS,GAAG,GAAG,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAgB,CAAA;wBACjD,IAAM,cAAc,GAAG,EAAE,CAAC,MAAM,CAAC,SAAS,EAAE,KAAI,CAAC,OAAO,CAAC,EAAE,CAAC,CAAA;wBAE5D,OAAO,cAAc,CAAA;oBACvB,CAAC,CAAC,EAAA;;;KACH;IACY,oCAAO,GAApB,UAAqB,KAAgB;;;;;;wBAC5B,KAAA,IAAI,CAAC,YAAY,CAAA;wBAAC,qBAAM,uBAAU,CAAC,KAAK,EAAE,IAAI,CAAC,EAAA;4BAAtD,sBAAO,SAAA,IAAI,GAAc,SAA6B,EAAC,EAAA;;;;KACxD;IAEY,kDAAqB,GAAlC,UAAmC,KAAgB;;;;;;wBAC5B,KAAA,IAAI,CAAC,OAAO,CAAA;wBAAC,qBAAM,uBAAU,CAAC,KAAK,EAAE,IAAI,CAAC,EAAA;4BAAhD,qBAAM,SAAA,IAAI,GAAS,SAA6B,EAAC,EAAA;;wBAA1D,MAAM,GAAG,SAAiD;wBACnD,qBAAM,MAAM,CAAC,IAAI,EAAE,EAAA;;wBAA1B,IAAI,GAAG,SAAmB;wBAChC,MAAM,CAAC,OAAO,EAAE,CAAA;wBAChB,sBAAO,IAAoB,EAAA;;;;KAC5B;IACH,yBAAC;AAAD,CAAC,AApED,IAoEC;AApEY,gDAAkB"}
\ No newline at end of file

build/getImageTensor.d.ts

-import * as tf from '@tensorflow/tfjs-core';
-import { NetInput } from './NetInput';
-import { TNetInput } from './types';
-export declare function getImageTensor(input: tf.Tensor | NetInput | TNetInput): tf.Tensor4D;

build/getImageTensor.js

-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-var tf = require("@tensorflow/tfjs-core");
-var NetInput_1 = require("./NetInput");
-function getImageTensor(input) {
-    return tf.tidy(function () {
-        if (input instanceof tf.Tensor) {
-            var rank = input.shape.length;
-            if (rank !== 3 && rank !== 4) {
-                throw new Error('input tensor must be of rank 3 or 4');
-            }
-            return (rank === 3 ? input.expandDims(0) : input).toFloat();
-        }
-        var netInput = input instanceof NetInput_1.NetInput ? input : new NetInput_1.NetInput(input);
-        return tf.concat(netInput.canvases.map(function (canvas) {
-            return tf.fromPixels(canvas).expandDims(0).toFloat();
-        }));
-    });
-}
-exports.getImageTensor = getImageTensor;
-//# sourceMappingURL=getImageTensor.js.map
\ No newline at end of file

build/getImageTensor.js.map

-{"version":3,"file":"getImageTensor.js","sourceRoot":"","sources":["../src/getImageTensor.ts"],"names":[],"mappings":";;AAAA,0CAA4C;AAE5C,uCAAsC;AAGtC,wBAA+B,KAAuC;IACpE,OAAO,EAAE,CAAC,IAAI,CAAC;QACb,IAAI,KAAK,YAAY,EAAE,CAAC,MAAM,EAAE;YAC9B,IAAM,IAAI,GAAG,KAAK,CAAC,KAAK,CAAC,MAAM,CAAA;YAC/B,IAAI,IAAI,KAAK,CAAC,IAAI,IAAI,KAAK,CAAC,EAAE;gBAC5B,MAAM,IAAI,KAAK,CAAC,qCAAqC,CAAC,CAAA;aACvD;YAED,OAAO,CAAC,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,OAAO,EAAiB,CAAA;SAC3E;QAED,IAAM,QAAQ,GAAG,KAAK,YAAY,mBAAQ,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,mBAAQ,CAAC,KAAK,CAAC,CAAA;QACxE,OAAO,EAAE,CAAC,MAAM,CACd,QAAQ,CAAC,QAAQ,CAAC,GAAG,CAAC,UAAA,MAAM;YAC1B,OAAA,EAAE,CAAC,UAAU,CAAC,MAAM,CAAC,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC,OAAO,EAAE;QAA7C,CAA6C,CAC9C,CACa,CAAA;IAClB,CAAC,CAAC,CAAA;AACJ,CAAC;AAlBD,wCAkBC"}
\ No newline at end of file

build/globalApi.d.ts

@@ -4,9 +4,9 @@ import { FaceDetectionNet } from './faceDetectionNet/FaceDetectionNet';
 import { FaceLandmarkNet } from './faceLandmarkNet/FaceLandmarkNet';
 import { FaceLandmarks } from './faceLandmarkNet/FaceLandmarks';
 import { FaceRecognitionNet } from './faceRecognitionNet/FaceRecognitionNet';
+import { FullFaceDescription } from './FullFaceDescription';
 import { NetInput } from './NetInput';
 import { TNetInput } from './types';
-import { FullFaceDescription } from './FullFaceDescription';
 export declare const detectionNet: FaceDetectionNet;
 export declare const landmarkNet: FaceLandmarkNet;
 export declare const recognitionNet: FaceRecognitionNet;
@@ -14,7 +14,7 @@ export declare function loadFaceDetectionModel(url: string): Promise<void>;
 export declare function loadFaceLandmarkModel(url: string): Promise<void>;
 export declare function loadFaceRecognitionModel(url: string): Promise<void>;
 export declare function loadModels(url: string): Promise<[void, void, void]>;
-export declare function locateFaces(input: tf.Tensor | NetInput | TNetInput, minConfidence?: number, maxResults?: number): Promise<FaceDetection[]>;
-export declare function detectLandmarks(input: tf.Tensor | NetInput | TNetInput): Promise<FaceLandmarks>;
-export declare function computeFaceDescriptor(input: tf.Tensor | NetInput | TNetInput): Promise<Float32Array>;
+export declare function locateFaces(input: TNetInput, minConfidence?: number, maxResults?: number): Promise<FaceDetection[]>;
+export declare function detectLandmarks(input: TNetInput): Promise<FaceLandmarks | FaceLandmarks[]>;
+export declare function computeFaceDescriptor(input: TNetInput): Promise<Float32Array>;
 export declare const allFaces: (input: tf.Tensor | NetInput | TNetInput, minConfidence: number) => Promise<FullFaceDescription[]>;

build/globalApi.js.map

-{"version":3,"file":"globalApi.js","sourceRoot":"","sources":["../src/globalApi.ts"],"names":[],"mappings":";;AAEA,qDAAoD;AAEpD,wEAAuE;AACvE,qEAAoE;AAEpE,8EAA6E;AAKhE,QAAA,YAAY,GAAG,IAAI,mCAAgB,EAAE,CAAA;AACrC,QAAA,WAAW,GAAG,IAAI,iCAAe,EAAE,CAAA;AACnC,QAAA,cAAc,GAAG,IAAI,uCAAkB,EAAE,CAAA;AAEtD,gCAAuC,GAAW;IAChD,OAAO,oBAAY,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;AAC/B,CAAC;AAFD,wDAEC;AAED,+BAAsC,GAAW;IAC/C,OAAO,mBAAW,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;AAC9B,CAAC;AAFD,sDAEC;AAED,kCAAyC,GAAW;IAClD,OAAO,sBAAc,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;AACjC,CAAC;AAFD,4DAEC;AAED,oBAA2B,GAAW;IACpC,OAAO,OAAO,CAAC,GAAG,CAAC;QACjB,sBAAsB,CAAC,GAAG,CAAC;QAC3B,qBAAqB,CAAC,GAAG,CAAC;QAC1B,wBAAwB,CAAC,GAAG,CAAC;KAC9B,CAAC,CAAA;AACJ,CAAC;AAND,gCAMC;AAED,qBACE,KAAuC,EACvC,aAAsB,EACtB,UAAmB;IAEnB,OAAO,oBAAY,CAAC,WAAW,CAAC,KAAK,EAAE,aAAa,EAAE,UAAU,CAAC,CAAA;AACnE,CAAC;AAND,kCAMC;AAED,yBACE,KAAuC;IAEvC,OAAO,mBAAW,CAAC,eAAe,CAAC,KAAK,CAAC,CAAA;AAC3C,CAAC;AAJD,0CAIC;AAED,+BACE,KAAuC;IAEvC,OAAO,sBAAc,CAAC,qBAAqB,CAAC,KAAK,CAAC,CAAA;AACpD,CAAC;AAJD,sDAIC;AAEY,QAAA,QAAQ,GAGiB,iCAAe,CACnD,oBAAY,EACZ,mBAAW,EACX,sBAAc,CACf,CAAA"}
\ No newline at end of file
+{"version":3,"file":"globalApi.js","sourceRoot":"","sources":["../src/globalApi.ts"],"names":[],"mappings":";;AAEA,qDAAoD;AAEpD,wEAAuE;AACvE,qEAAoE;AAEpE,8EAA6E;AAKhE,QAAA,YAAY,GAAG,IAAI,mCAAgB,EAAE,CAAA;AACrC,QAAA,WAAW,GAAG,IAAI,iCAAe,EAAE,CAAA;AACnC,QAAA,cAAc,GAAG,IAAI,uCAAkB,EAAE,CAAA;AAEtD,gCAAuC,GAAW;IAChD,OAAO,oBAAY,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;AAC/B,CAAC;AAFD,wDAEC;AAED,+BAAsC,GAAW;IAC/C,OAAO,mBAAW,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;AAC9B,CAAC;AAFD,sDAEC;AAED,kCAAyC,GAAW;IAClD,OAAO,sBAAc,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;AACjC,CAAC;AAFD,4DAEC;AAED,oBAA2B,GAAW;IACpC,OAAO,OAAO,CAAC,GAAG,CAAC;QACjB,sBAAsB,CAAC,GAAG,CAAC;QAC3B,qBAAqB,CAAC,GAAG,CAAC;QAC1B,wBAAwB,CAAC,GAAG,CAAC;KAC9B,CAAC,CAAA;AACJ,CAAC;AAND,gCAMC;AAED,qBACE,KAAgB,EAChB,aAAsB,EACtB,UAAmB;IAEnB,OAAO,oBAAY,CAAC,WAAW,CAAC,KAAK,EAAE,aAAa,EAAE,UAAU,CAAC,CAAA;AACnE,CAAC;AAND,kCAMC;AAED,yBACE,KAAgB;IAEhB,OAAO,mBAAW,CAAC,eAAe,CAAC,KAAK,CAAC,CAAA;AAC3C,CAAC;AAJD,0CAIC;AAED,+BACE,KAAgB;IAEhB,OAAO,sBAAc,CAAC,qBAAqB,CAAC,KAAK,CAAC,CAAA;AACpD,CAAC;AAJD,sDAIC;AAEY,QAAA,QAAQ,GAGiB,iCAAe,CACnD,oBAAY,EACZ,mBAAW,EACX,sBAAc,CACf,CAAA"}
\ No newline at end of file

build/padToSquare.js.map

-{"version":3,"file":"padToSquare.js","sourceRoot":"","sources":["../src/padToSquare.ts"],"names":[],"mappings":";;AAAA,0CAA4C;AAI5C;;;;;;GAMG;AACH,qBACE,SAAsB,EACtB,aAA8B;IAA9B,8BAAA,EAAA,qBAA8B;IAE9B,OAAO,EAAE,CAAC,IAAI,CAAC;QAEP,IAAA,6BAA0C,EAAzC,cAAM,EAAE,aAAK,CAA4B;QAChD,IAAI,MAAM,KAAK,KAAK,EAAE;YACpB,OAAO,SAAS,CAAA;SACjB;QAED,IAAM,OAAO,GAAG,IAAI,CAAC,GAAG,CAAC,MAAM,GAAG,KAAK,CAAC,CAAA;QACxC,IAAM,aAAa,GAAG,IAAI,CAAC,KAAK,CAAC,OAAO,GAAG,CAAC,aAAa,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAA;QACrE,IAAM,WAAW,GAAG,MAAM,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAA;QAE1C,IAAM,mBAAmB,GAAG,UAAC,aAAqB;YAChD,IAAM,kBAAkB,GAAG,SAAS,CAAC,KAAK,CAAC,KAAK,EAAE,CAAA;YAClD,kBAAkB,CAAC,WAAW,CAAC,GAAG,aAAa,CAAA;YAC/C,OAAO,EAAE,CAAC,IAAI,CAAC,kBAAkB,EAAE,CAAC,CAAC,CAAA;QACvC,CAAC,CAAA;QAED,IAAM,mBAAmB,GAAG,mBAAmB,CAAC,aAAa,CAAC,CAAA;QAC9D,IAAM,sBAAsB,GAAG,OAAO,GAAG,mBAAmB,CAAC,KAAK,CAAC,WAAW,CAAC,CAAA;QAE/E,IAAM,oBAAoB,GAAG,aAAa,IAAI,sBAAsB;YAClE,CAAC,CAAC,mBAAmB,CAAC,sBAAsB,CAAC;YAC7C,CAAC,CAAC,IAAI,CAAA;QAER,IAAM,cAAc,GAAG;YACrB,oBAAoB;YACpB,SAAS;YACT,mBAAmB;SACpB;aACE,MAAM,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,KAAK,IAAI,EAAV,CAAU,CAAkB,CAAA;QAC3C,OAAO,EAAE,CAAC,MAAM,CAAC,cAAc,EAAE,WAAW,CAAC,CAAA;IAC/C,CAAC,CAAC,CAAA;AACJ,CAAC;AApCD,kCAoCC"}
\ No newline at end of file
+{"version":3,"file":"padToSquare.js","sourceRoot":"","sources":["../src/padToSquare.ts"],"names":[],"mappings":";;AAAA,0CAA4C;AAE5C;;;;;;GAMG;AACH,qBACE,SAAsB,EACtB,aAA8B;IAA9B,8BAAA,EAAA,qBAA8B;IAE9B,OAAO,EAAE,CAAC,IAAI,CAAC;QAEP,IAAA,6BAA0C,EAAzC,cAAM,EAAE,aAAK,CAA4B;QAChD,IAAI,MAAM,KAAK,KAAK,EAAE;YACpB,OAAO,SAAS,CAAA;SACjB;QAED,IAAM,OAAO,GAAG,IAAI,CAAC,GAAG,CAAC,MAAM,GAAG,KAAK,CAAC,CAAA;QACxC,IAAM,aAAa,GAAG,IAAI,CAAC,KAAK,CAAC,OAAO,GAAG,CAAC,aAAa,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAA;QACrE,IAAM,WAAW,GAAG,MAAM,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAA;QAE1C,IAAM,mBAAmB,GAAG,UAAC,aAAqB;YAChD,IAAM,kBAAkB,GAAG,SAAS,CAAC,KAAK,CAAC,KAAK,EAAE,CAAA;YAClD,kBAAkB,CAAC,WAAW,CAAC,GAAG,aAAa,CAAA;YAC/C,OAAO,EAAE,CAAC,IAAI,CAAC,kBAAkB,EAAE,CAAC,CAAC,CAAA;QACvC,CAAC,CAAA;QAED,IAAM,mBAAmB,GAAG,mBAAmB,CAAC,aAAa,CAAC,CAAA;QAC9D,IAAM,sBAAsB,GAAG,OAAO,GAAG,mBAAmB,CAAC,KAAK,CAAC,WAAW,CAAC,CAAA;QAE/E,IAAM,oBAAoB,GAAG,aAAa,IAAI,sBAAsB;YAClE,CAAC,CAAC,mBAAmB,CAAC,sBAAsB,CAAC;YAC7C,CAAC,CAAC,IAAI,CAAA;QAER,IAAM,cAAc,GAAG;YACrB,oBAAoB;YACpB,SAAS;YACT,mBAAmB;SACpB;aACE,MAAM,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,KAAK,IAAI,EAAV,CAAU,CAAkB,CAAA;QAC3C,OAAO,EAAE,CAAC,MAAM,CAAC,cAAc,EAAE,WAAW,CAAC,CAAA;IAC/C,CAAC,CAAC,CAAA;AACJ,CAAC;AApCD,kCAoCC"}
\ No newline at end of file

build/toNetInput.d.ts

@@ -5,6 +5,8 @@ import { TNetInput } from './types';
  * to be finished loading.
  *
  * @param input The input, which can be a media element or an array of different media elements.
+ * @param manageCreatedInput If a new NetInput instance is created from the inputs, this flag
+ * determines, whether to set the NetInput as managed or not.
  * @returns A NetInput instance, which can be passed into one of the neural networks.
  */
-export declare function toNetInput(input: NetInput | TNetInput): Promise<NetInput>;
+export declare function toNetInput(inputs: TNetInput, manageCreatedInput?: boolean): Promise<NetInput>;

build/toNetInput.js

 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
 var tslib_1 = require("tslib");
+var isMediaElement_1 = require("./commons/isMediaElement");
+var isTensor_1 = require("./commons/isTensor");
 var NetInput_1 = require("./NetInput");
 var utils_1 = require("./utils");
 /**
@@ -8,39 +10,61 @@ var utils_1 = require("./utils");
  * to be finished loading.
  *
  * @param input The input, which can be a media element or an array of different media elements.
+ * @param manageCreatedInput If a new NetInput instance is created from the inputs, this flag
+ * determines, whether to set the NetInput as managed or not.
  * @returns A NetInput instance, which can be passed into one of the neural networks.
  */
-function toNetInput(input) {
+function toNetInput(inputs, manageCreatedInput) {
+    if (manageCreatedInput === void 0) { manageCreatedInput = false; }
     return tslib_1.__awaiter(this, void 0, void 0, function () {
-        var mediaArgArray, medias;
+        var afterCreate, inputArgArray, getIdxHint, inputArray;
         return tslib_1.__generator(this, function (_a) {
             switch (_a.label) {
                 case 0:
-                    if (input instanceof NetInput_1.NetInput) {
-                        return [2 /*return*/, input];
+                    if (inputs instanceof NetInput_1.NetInput) {
+                        return [2 /*return*/, inputs];
                     }
-                    mediaArgArray = Array.isArray(input)
-                        ? input
-                        : [input];
-                    if (!mediaArgArray.length) {
+                    afterCreate = function (netInput) { return manageCreatedInput
+                        ? netInput.managed()
+                        : netInput; };
+                    if (isTensor_1.isTensor4D(inputs)) {
+                        return [2 /*return*/, afterCreate(new NetInput_1.NetInput(inputs))];
+                    }
+                    inputArgArray = Array.isArray(inputs)
+                        ? inputs
+                        : [inputs];
+                    if (!inputArgArray.length) {
                         throw new Error('toNetInput - empty array passed as input');
                     }
-                    medias = mediaArgArray.map(utils_1.getElement);
-                    medias.forEach(function (media, i) {
-                        if (!(media instanceof HTMLImageElement || media instanceof HTMLVideoElement || media instanceof HTMLCanvasElement)) {
-                            var idxHint = Array.isArray(input) ? " at input index " + i + ":" : '';
-                            if (typeof mediaArgArray[i] === 'string') {
-                                throw new Error("toNetInput -" + idxHint + " string passed, but could not resolve HTMLElement for element id");
+                    getIdxHint = function (idx) { return Array.isArray(inputs) ? " at input index " + idx + ":" : ''; };
+                    inputArray = inputArgArray
+                        .map(utils_1.resolveInput)
+                        .map(function (input, i) {
+                        if (isTensor_1.isTensor4D(input)) {
+                            // if tf.Tensor4D is passed in the input array, the batch size has to be 1
+                            var batchSize = input.shape[0];
+                            if (batchSize !== 1) {
+                                throw new Error("toNetInput -" + getIdxHint(i) + " tf.Tensor4D with batchSize " + batchSize + " passed, but not supported in input array");
+                            }
+                            // to tf.Tensor3D
+                            return input.reshape(input.shape.slice(1));
+                        }
+                        return input;
+                    });
+                    inputArray.forEach(function (input, i) {
+                        if (!isMediaElement_1.isMediaElement(input) && !isTensor_1.isTensor3D(input)) {
+                            if (typeof inputArgArray[i] === 'string') {
+                                throw new Error("toNetInput -" + getIdxHint(i) + " string passed, but could not resolve HTMLElement for element id " + inputArgArray[i]);
                             }
-                            throw new Error("toNetInput -" + idxHint + " expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement, or to be an element id");
+                            throw new Error("toNetInput -" + getIdxHint(i) + " expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement | tf.Tensor3D, or to be an element id");
                         }
                     });
                     // wait for all media elements being loaded
-                    return [4 /*yield*/, Promise.all(medias.map(function (media) { return utils_1.awaitMediaLoaded(media); }))];
+                    return [4 /*yield*/, Promise.all(inputArray.map(function (input) { return isMediaElement_1.isMediaElement(input) && utils_1.awaitMediaLoaded(input); }))];
                 case 1:
                     // wait for all media elements being loaded
                     _a.sent();
-                    return [2 /*return*/, new NetInput_1.NetInput(medias)];
+                    return [2 /*return*/, afterCreate(new NetInput_1.NetInput(inputArray, Array.isArray(inputs)))];
             }
         });
     });

build/toNetInput.js.map

-{"version":3,"file":"toNetInput.js","sourceRoot":"","sources":["../src/toNetInput.ts"],"names":[],"mappings":";;;AAAA,uCAAsC;AAEtC,iCAAuD;AAEvD;;;;;;GAMG;AACH,oBACE,KAA2B;;;;;;oBAE3B,IAAI,KAAK,YAAY,mBAAQ,EAAE;wBAC7B,sBAAO,KAAK,EAAA;qBACb;oBAEK,aAAa,GAAG,KAAK,CAAC,OAAO,CAAC,KAAK,CAAC;wBACtC,CAAC,CAAC,KAAK;wBACP,CAAC,CAAC,CAAC,KAAK,CAAC,CAAA;oBAEb,IAAI,CAAC,aAAa,CAAC,MAAM,EAAE;wBACzB,MAAM,IAAI,KAAK,CAAC,0CAA0C,CAAC,CAAA;qBAC5D;oBAEK,MAAM,GAAG,aAAa,CAAC,GAAG,CAAC,kBAAU,CAAC,CAAA;oBAE5C,MAAM,CAAC,OAAO,CAAC,UAAC,KAAK,EAAE,CAAC;wBACtB,IAAI,CAAC,CAAC,KAAK,YAAY,gBAAgB,IAAI,KAAK,YAAY,gBAAgB,IAAI,KAAK,YAAY,iBAAiB,CAAC,EAAE;4BACnH,IAAM,OAAO,GAAG,KAAK,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,qBAAmB,CAAC,MAAG,CAAC,CAAC,CAAC,EAAE,CAAA;4BACnE,IAAI,OAAO,aAAa,CAAC,CAAC,CAAC,KAAK,QAAQ,EAAE;gCACxC,MAAM,IAAI,KAAK,CAAC,iBAAe,OAAO,qEAAkE,CAAC,CAAA;6BAC1G;4BACD,MAAM,IAAI,KAAK,CAAC,iBAAe,OAAO,kHAA+G,CAAC,CAAA;yBACvJ;oBACH,CAAC,CAAC,CAAA;oBAEF,2CAA2C;oBAC3C,qBAAM,OAAO,CAAC,GAAG,CACf,MAAM,CAAC,GAAG,CAAC,UAAA,KAAK,IAAI,OAAA,wBAAgB,CAAC,KAAK,CAAC,EAAvB,CAAuB,CAAC,CAC7C,EAAA;;oBAHD,2CAA2C;oBAC3C,SAEC,CAAA;oBAED,sBAAO,IAAI,mBAAQ,CAAC,MAAM,CAAC,EAAA;;;;CAC5B;AAjCD,gCAiCC"}
\ No newline at end of file
+{"version":3,"file":"toNetInput.js","sourceRoot":"","sources":["../src/toNetInput.ts"],"names":[],"mappings":";;;AAEA,2DAA0D;AAC1D,+CAA4D;AAC5D,uCAAsC;AAEtC,iCAAyD;AAEzD;;;;;;;;GAQG;AACH,oBACE,MAAiB,EACjB,kBAAmC;IAAnC,mCAAA,EAAA,0BAAmC;;;;;;oBAEnC,IAAI,MAAM,YAAY,mBAAQ,EAAE;wBAC9B,sBAAO,MAAM,EAAA;qBACd;oBAEK,WAAW,GAAG,UAAC,QAAkB,IAAK,OAAA,kBAAkB;wBAC5D,CAAC,CAAC,QAAQ,CAAC,OAAO,EAAE;wBACpB,CAAC,CAAC,QAAQ,EAFgC,CAEhC,CAAA;oBAEZ,IAAI,qBAAU,CAAC,MAAM,CAAC,EAAE;wBACtB,sBAAO,WAAW,CAAC,IAAI,mBAAQ,CAAC,MAAqB,CAAC,CAAC,EAAA;qBACxD;oBAEG,aAAa,GAAG,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC;wBACrC,CAAC,CAAC,MAAM;wBACR,CAAC,CAAC,CAAC,MAAM,CAAC,CAAA;oBAEd,IAAI,CAAC,aAAa,CAAC,MAAM,EAAE;wBACzB,MAAM,IAAI,KAAK,CAAC,0CAA0C,CAAC,CAAA;qBAC5D;oBAEK,UAAU,GAAG,UAAC,GAAW,IAAK,OAAA,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,qBAAmB,GAAG,MAAG,CAAC,CAAC,CAAC,EAAE,EAAtD,CAAsD,CAAA;oBAEpF,UAAU,GAAG,aAAa;yBAC7B,GAAG,CAAC,oBAAY,CAAC;yBACjB,GAAG,CAAC,UAAC,KAAK,EAAE,CAAC;wBACZ,IAAI,qBAAU,CAAC,KAAK,CAAC,EAAE;4BACrB,0EAA0E;4BAC1E,IAAM,SAAS,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC,CAAA;4BAChC,IAAI,SAAS,KAAK,CAAC,EAAE;gCACnB,MAAM,IAAI,KAAK,CAAC,iBAAe,UAAU,CAAC,CAAC,CAAC,oCAA+B,SAAS,8CAA2C,CAAC,CAAA;6BACjI;4BACD,iBAAiB;4BACjB,OAAO,KAAK,CAAC,OAAO,CAAC,KAAK,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAA;yBAC3C;wBACD,OAAO,KAAK,CAAA;oBACd,CAAC,CAAC,CAAA;oBAEJ,UAAU,CAAC,OAAO,CAAC,UAAC,KAAK,EAAE,CAAC;wBAC1B,IAAI,CAAC,+BAAc,CAAC,KAAK,CAAC,IAAI,CAAC,qBAAU,CAAC,KAAK,CAAC,EAAE;4BAEhD,IAAI,OAAO,aAAa,CAAC,CAAC,CAAC,KAAK,QAAQ,EAAE;gCACxC,MAAM,IAAI,KAAK,CAAC,iBAAe,UAAU,CAAC,CAAC,CAAC,yEAAoE,aAAa,CAAC,CAAC,CAAG,CAAC,CAAA;6BACpI;4BAED,MAAM,IAAI,KAAK,CAAC,iBAAe,UAAU,CAAC,CAAC,CAAC,gIAA6H,CAAC,CAAA;yBAC3K;oBACH,CAAC,CAAC,CAAA;oBAEF,2CAA2C;oBAC3C,qBAAM,OAAO,CAAC,GAAG,CACf,UAAU,CAAC,GAAG,CAAC,UAAA,KAAK,IAAI,OAAA,+BAAc,CAAC,KAAK,CAAC,IAAI,wBAAgB,CAAC,KAAK,CAAC,EAAhD,CAAgD,CAAC,CAC1E,EAAA;;oBAHD,2CAA2C;oBAC3C,SAEC,CAAA;oBAED,sBAAO,WAAW,CAAC,IAAI,mBAAQ,CAAC,UAAU,EAAE,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC,CAAC,EAAA;;;;CACpE;AA1DD,gCA0DC"}
\ No newline at end of file

build/types.d.ts

+import * as tf from '@tensorflow/tfjs-core';
+import { NetInput } from './NetInput';
 export declare type TMediaElement = HTMLImageElement | HTMLVideoElement | HTMLCanvasElement;
-export declare type TNetInputArg = string | TMediaElement;
-export declare type TNetInput = TNetInputArg | Array<TNetInputArg>;
+export declare type TResolvedNetInput = TMediaElement | tf.Tensor3D | tf.Tensor4D;
+export declare type TNetInputArg = string | TResolvedNetInput;
+export declare type TNetInput = TNetInputArg | Array<TNetInputArg> | NetInput | tf.Tensor4D;
 export declare type Dimensions = {
     width: number;
     height: number;

build/utils.d.ts

@@ -3,7 +3,7 @@ import { Dimensions } from './types';
 export declare function isFloat(num: number): boolean;
 export declare function isEven(num: number): boolean;
 export declare function round(num: number): number;
-export declare function getElement(arg: string | any): any;
+export declare function resolveInput(arg: string | any): any;
 export declare function isLoaded(media: HTMLImageElement | HTMLVideoElement): boolean;
 export declare function awaitMediaLoaded(media: HTMLImageElement | HTMLVideoElement | HTMLCanvasElement): Promise<{}>;
 export declare function getContext2dOrThrow(canvas: HTMLCanvasElement): CanvasRenderingContext2D;
@@ -14,4 +14,4 @@ export declare function getMediaDimensions(media: HTMLImageElement | HTMLVideoEl
     height: number;
 };
 export declare function bufferToImage(buf: Blob): Promise<HTMLImageElement>;
-export declare function imageTensorToCanvas(imgTensor: tf.Tensor4D, canvas?: HTMLCanvasElement): Promise<HTMLCanvasElement>;
+export declare function imageTensorToCanvas(imgTensor: tf.Tensor, canvas?: HTMLCanvasElement): Promise<HTMLCanvasElement>;

build/utils.js

@@ -2,6 +2,7 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 var tslib_1 = require("tslib");
 var tf = require("@tensorflow/tfjs-core");
+var isTensor_1 = require("./commons/isTensor");
 function isFloat(num) {
     return num % 1 !== 0;
 }
@@ -14,13 +15,13 @@ function round(num) {
     return Math.floor(num * 100) / 100;
 }
 exports.round = round;
-function getElement(arg) {
+function resolveInput(arg) {
     if (typeof arg === 'string') {
         return document.getElementById(arg);
     }
     return arg;
 }
-exports.getElement = getElement;
+exports.resolveInput = resolveInput;
 function isLoaded(media) {
     return (media instanceof HTMLImageElement && media.complete)
         || (media instanceof HTMLVideoElement && media.readyState >= 3);
@@ -105,12 +106,12 @@ function bufferToImage(buf) {
 exports.bufferToImage = bufferToImage;
 function imageTensorToCanvas(imgTensor, canvas) {
     return tslib_1.__awaiter(this, void 0, void 0, function () {
-        var targetCanvas, _a, _, height, width, numChannels;
+        var targetCanvas, _a, height, width, numChannels;
         return tslib_1.__generator(this, function (_b) {
             switch (_b.label) {
                 case 0:
                     targetCanvas = canvas || document.createElement('canvas');
-                    _a = imgTensor.shape, _ = _a[0], height = _a[1], width = _a[2], numChannels = _a[3];
+                    _a = imgTensor.shape.slice(isTensor_1.isTensor4D(imgTensor) ? 1 : 0), height = _a[0], width = _a[1], numChannels = _a[2];
                     return [4 /*yield*/, tf.toPixels(imgTensor.as3D(height, width, numChannels).toInt(), targetCanvas)];
                 case 1:
                     _b.sent();

build/utils.js.map

-{"version":3,"file":"utils.js","sourceRoot":"","sources":["../src/utils.ts"],"names":[],"mappings":";;;AAAA,0CAA4C;AAI5C,iBAAwB,GAAW;IACjC,OAAO,GAAG,GAAG,CAAC,KAAK,CAAC,CAAA;AACtB,CAAC;AAFD,0BAEC;AAED,gBAAuB,GAAW;IAChC,OAAO,GAAG,GAAG,CAAC,KAAK,CAAC,CAAA;AACtB,CAAC;AAFD,wBAEC;AAED,eAAsB,GAAW;IAC/B,OAAO,IAAI,CAAC,KAAK,CAAC,GAAG,GAAG,GAAG,CAAC,GAAG,GAAG,CAAA;AACpC,CAAC;AAFD,sBAEC;AAED,oBAA2B,GAAiB;IAC1C,IAAI,OAAO,GAAG,KAAK,QAAQ,EAAE;QAC3B,OAAO,QAAQ,CAAC,cAAc,CAAC,GAAG,CAAC,CAAA;KACpC;IACD,OAAO,GAAG,CAAA;AACZ,CAAC;AALD,gCAKC;AAED,kBAAyB,KAA0C;IACjE,OAAO,CAAC,KAAK,YAAY,gBAAgB,IAAI,KAAK,CAAC,QAAQ,CAAC;WACvD,CAAC,KAAK,YAAY,gBAAgB,IAAI,KAAK,CAAC,UAAU,IAAI,CAAC,CAAC,CAAA;AACnE,CAAC;AAHD,4BAGC;AAED,0BAAiC,KAA8D;IAC7F,OAAO,IAAI,OAAO,CAAC,UAAC,OAAO,EAAE,MAAM;QACjC,IAAI,KAAK,YAAY,iBAAiB,IAAI,QAAQ,CAAC,KAAK,CAAC,EAAE;YACzD,OAAO,OAAO,EAAE,CAAA;SACjB;QAED,gBAAgB,CAAQ;YACtB,IAAI,CAAC,CAAC,CAAC,aAAa;gBAAE,OAAM;YAC5B,CAAC,CAAC,aAAa,CAAC,mBAAmB,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;YACnD,CAAC,CAAC,aAAa,CAAC,mBAAmB,CAAC,OAAO,EAAE,OAAO,CAAC,CAAA;YACrD,OAAO,EAAE,CAAA;QACX,CAAC;QAED,iBAAiB,CAAQ;YACvB,IAAI,CAAC,CAAC,CAAC,aAAa;gBAAE,OAAM;YAC5B,CAAC,CAAC,aAAa,CAAC,mBAAmB,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;YACnD,CAAC,CAAC,aAAa,CAAC,mBAAmB,CAAC,OAAO,EAAE,OAAO,CAAC,CAAA;YACrD,MAAM,EAAE,CAAA;QACV,CAAC;QAED,KAAK,CAAC,gBAAgB,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;QACtC,KAAK,CAAC,gBAAgB,CAAC,OAAO,EAAE,OAAO,CAAC,CAAA;IAC1C,CAAC,CAAC,CAAA;AACJ,CAAC;AAvBD,4CAuBC;AAED,6BAAoC,MAAyB;IAC3D,IAAM,GAAG,GAAG,MAAM,CAAC,UAAU,CAAC,IAAI,CAAC,CAAA;IACnC,IAAI,CAAC,GAAG,EAAE;QACR,MAAM,IAAI,KAAK,CAAC,2BAA2B,CAAC,CAAA;KAC7C;IACD,OAAO,GAAG,CAAA;AACZ,CAAC;AAND,kDAMC;AAED,sBAA6B,EAA6B;QAA3B,gBAAK,EAAE,kBAAM;IAC1C,IAAM,MAAM,GAAG,QAAQ,CAAC,aAAa,CAAC,QAAQ,CAAC,CAAA;IAC/C,MAAM,CAAC,KAAK,GAAG,KAAK,CAAA;IACpB,MAAM,CAAC,MAAM,GAAG,MAAM,CAAA;IACtB,OAAO,MAAM,CAAA;AACf,CAAC;AALD,oCAKC;AAED,+BAAsC,KAA0C,EAAE,IAAiB;IACjG,IAAI,CAAC,QAAQ,CAAC,KAAK,CAAC,EAAE;QACpB,MAAM,IAAI,KAAK,CAAC,4DAA4D,CAAC,CAAA;KAC9E;IAEK,IAAA,sCAAqD,EAAnD,gBAAK,EAAE,kBAAM,CAAsC;IAC3D,IAAM,MAAM,GAAG,YAAY,CAAC,EAAE,KAAK,OAAA,EAAE,MAAM,QAAA,EAAE,CAAC,CAAA;IAC9C,mBAAmB,CAAC,MAAM,CAAC,CAAC,SAAS,CAAC,KAAK,EAAE,CAAC,EAAE,CAAC,EAAE,KAAK,EAAE,MAAM,CAAC,CAAA;IACjE,OAAO,MAAM,CAAA;AACf,CAAC;AATD,sDASC;AAED,4BAAmC,KAA0C;IAC3E,IAAI,KAAK,YAAY,gBAAgB,EAAE;QACrC,OAAO,EAAE,KAAK,EAAE,KAAK,CAAC,YAAY,EAAE,MAAM,EAAE,KAAK,CAAC,aAAa,EAAE,CAAA;KAClE;IACD,IAAI,KAAK,YAAY,gBAAgB,EAAE;QACrC,OAAO,EAAE,KAAK,EAAE,KAAK,CAAC,UAAU,EAAE,MAAM,EAAE,KAAK,CAAC,WAAW,EAAE,CAAA;KAC9D;IACD,OAAO,KAAK,CAAA;AACd,CAAC;AARD,gDAQC;AAED,uBAA8B,GAAS;IACrC,OAAO,IAAI,OAAO,CAAC,UAAC,OAAO,EAAE,MAAM;QACjC,IAAI,CAAC,CAAC,GAAG,YAAY,IAAI,CAAC,EAAE;YAC1B,OAAO,MAAM,CAAC,kDAAkD,CAAC,CAAA;SAClE;QAED,IAAM,MAAM,GAAG,IAAI,UAAU,EAAE,CAAA;QAC/B,MAAM,CAAC,MAAM,GAAG;YACd,IAAM,GAAG,GAAG,IAAI,KAAK,EAAE,CAAA;YACvB,GAAG,CAAC,MAAM,GAAG,cAAM,OAAA,OAAO,CAAC,GAAG,CAAC,EAAZ,CAAY,CAAA;YAC/B,GAAG,CAAC,OAAO,GAAG,MAAM,CAAA;YACpB,GAAG,CAAC,GAAG,GAAG,MAAM,CAAC,MAAM,CAAA;QACzB,CAAC,CAAA;QACD,MAAM,CAAC,OAAO,GAAG,MAAM,CAAA;QACvB,MAAM,CAAC,aAAa,CAAC,GAAG,CAAC,CAAA;IAC3B,CAAC,CAAC,CAAA;AACJ,CAAC;AAhBD,sCAgBC;AAED,6BACE,SAAsB,EACtB,MAA0B;;;;;;oBAEpB,YAAY,GAAG,MAAM,IAAK,QAAQ,CAAC,aAAa,CAAC,QAAQ,CAAC,CAAA;oBAE1D,KAAkC,SAAS,CAAC,KAAK,EAAhD,CAAC,QAAA,EAAE,MAAM,QAAA,EAAE,KAAK,QAAA,EAAE,WAAW,QAAA,CAAmB;oBACvD,qBAAM,EAAE,CAAC,QAAQ,CAAC,SAAS,CAAC,IAAI,CAAC,MAAM,EAAE,KAAK,EAAE,WAAW,CAAC,CAAC,KAAK,EAAE,EAAE,YAAY,CAAC,EAAA;;oBAAnF,SAAmF,CAAA;oBAEnF,sBAAO,YAAY,EAAA;;;;CACpB;AAVD,kDAUC"}
\ No newline at end of file
+{"version":3,"file":"utils.js","sourceRoot":"","sources":["../src/utils.ts"],"names":[],"mappings":";;;AAAA,0CAA4C;AAE5C,+CAAgD;AAGhD,iBAAwB,GAAW;IACjC,OAAO,GAAG,GAAG,CAAC,KAAK,CAAC,CAAA;AACtB,CAAC;AAFD,0BAEC;AAED,gBAAuB,GAAW;IAChC,OAAO,GAAG,GAAG,CAAC,KAAK,CAAC,CAAA;AACtB,CAAC;AAFD,wBAEC;AAED,eAAsB,GAAW;IAC/B,OAAO,IAAI,CAAC,KAAK,CAAC,GAAG,GAAG,GAAG,CAAC,GAAG,GAAG,CAAA;AACpC,CAAC;AAFD,sBAEC;AAED,sBAA6B,GAAiB;IAC5C,IAAI,OAAO,GAAG,KAAK,QAAQ,EAAE;QAC3B,OAAO,QAAQ,CAAC,cAAc,CAAC,GAAG,CAAC,CAAA;KACpC;IACD,OAAO,GAAG,CAAA;AACZ,CAAC;AALD,oCAKC;AAED,kBAAyB,KAA0C;IACjE,OAAO,CAAC,KAAK,YAAY,gBAAgB,IAAI,KAAK,CAAC,QAAQ,CAAC;WACvD,CAAC,KAAK,YAAY,gBAAgB,IAAI,KAAK,CAAC,UAAU,IAAI,CAAC,CAAC,CAAA;AACnE,CAAC;AAHD,4BAGC;AAED,0BAAiC,KAA8D;IAC7F,OAAO,IAAI,OAAO,CAAC,UAAC,OAAO,EAAE,MAAM;QACjC,IAAI,KAAK,YAAY,iBAAiB,IAAI,QAAQ,CAAC,KAAK,CAAC,EAAE;YACzD,OAAO,OAAO,EAAE,CAAA;SACjB;QAED,gBAAgB,CAAQ;YACtB,IAAI,CAAC,CAAC,CAAC,aAAa;gBAAE,OAAM;YAC5B,CAAC,CAAC,aAAa,CAAC,mBAAmB,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;YACnD,CAAC,CAAC,aAAa,CAAC,mBAAmB,CAAC,OAAO,EAAE,OAAO,CAAC,CAAA;YACrD,OAAO,EAAE,CAAA;QACX,CAAC;QAED,iBAAiB,CAAQ;YACvB,IAAI,CAAC,CAAC,CAAC,aAAa;gBAAE,OAAM;YAC5B,CAAC,CAAC,aAAa,CAAC,mBAAmB,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;YACnD,CAAC,CAAC,aAAa,CAAC,mBAAmB,CAAC,OAAO,EAAE,OAAO,CAAC,CAAA;YACrD,MAAM,EAAE,CAAA;QACV,CAAC;QAED,KAAK,CAAC,gBAAgB,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;QACtC,KAAK,CAAC,gBAAgB,CAAC,OAAO,EAAE,OAAO,CAAC,CAAA;IAC1C,CAAC,CAAC,CAAA;AACJ,CAAC;AAvBD,4CAuBC;AAED,6BAAoC,MAAyB;IAC3D,IAAM,GAAG,GAAG,MAAM,CAAC,UAAU,CAAC,IAAI,CAAC,CAAA;IACnC,IAAI,CAAC,GAAG,EAAE;QACR,MAAM,IAAI,KAAK,CAAC,2BAA2B,CAAC,CAAA;KAC7C;IACD,OAAO,GAAG,CAAA;AACZ,CAAC;AAND,kDAMC;AAED,sBAA6B,EAA6B;QAA3B,gBAAK,EAAE,kBAAM;IAC1C,IAAM,MAAM,GAAG,QAAQ,CAAC,aAAa,CAAC,QAAQ,CAAC,CAAA;IAC/C,MAAM,CAAC,KAAK,GAAG,KAAK,CAAA;IACpB,MAAM,CAAC,MAAM,GAAG,MAAM,CAAA;IACtB,OAAO,MAAM,CAAA;AACf,CAAC;AALD,oCAKC;AAED,+BAAsC,KAA0C,EAAE,IAAiB;IACjG,IAAI,CAAC,QAAQ,CAAC,KAAK,CAAC,EAAE;QACpB,MAAM,IAAI,KAAK,CAAC,4DAA4D,CAAC,CAAA;KAC9E;IAEK,IAAA,sCAAqD,EAAnD,gBAAK,EAAE,kBAAM,CAAsC;IAC3D,IAAM,MAAM,GAAG,YAAY,CAAC,EAAE,KAAK,OAAA,EAAE,MAAM,QAAA,EAAE,CAAC,CAAA;IAC9C,mBAAmB,CAAC,MAAM,CAAC,CAAC,SAAS,CAAC,KAAK,EAAE,CAAC,EAAE,CAAC,EAAE,KAAK,EAAE,MAAM,CAAC,CAAA;IACjE,OAAO,MAAM,CAAA;AACf,CAAC;AATD,sDASC;AAED,4BAAmC,KAA0C;IAC3E,IAAI,KAAK,YAAY,gBAAgB,EAAE;QACrC,OAAO,EAAE,KAAK,EAAE,KAAK,CAAC,YAAY,EAAE,MAAM,EAAE,KAAK,CAAC,aAAa,EAAE,CAAA;KAClE;IACD,IAAI,KAAK,YAAY,gBAAgB,EAAE;QACrC,OAAO,EAAE,KAAK,EAAE,KAAK,CAAC,UAAU,EAAE,MAAM,EAAE,KAAK,CAAC,WAAW,EAAE,CAAA;KAC9D;IACD,OAAO,KAAK,CAAA;AACd,CAAC;AARD,gDAQC;AAED,uBAA8B,GAAS;IACrC,OAAO,IAAI,OAAO,CAAC,UAAC,OAAO,EAAE,MAAM;QACjC,IAAI,CAAC,CAAC,GAAG,YAAY,IAAI,CAAC,EAAE;YAC1B,OAAO,MAAM,CAAC,kDAAkD,CAAC,CAAA;SAClE;QAED,IAAM,MAAM,GAAG,IAAI,UAAU,EAAE,CAAA;QAC/B,MAAM,CAAC,MAAM,GAAG;YACd,IAAM,GAAG,GAAG,IAAI,KAAK,EAAE,CAAA;YACvB,GAAG,CAAC,MAAM,GAAG,cAAM,OAAA,OAAO,CAAC,GAAG,CAAC,EAAZ,CAAY,CAAA;YAC/B,GAAG,CAAC,OAAO,GAAG,MAAM,CAAA;YACpB,GAAG,CAAC,GAAG,GAAG,MAAM,CAAC,MAAM,CAAA;QACzB,CAAC,CAAA;QACD,MAAM,CAAC,OAAO,GAAG,MAAM,CAAA;QACvB,MAAM,CAAC,aAAa,CAAC,GAAG,CAAC,CAAA;IAC3B,CAAC,CAAC,CAAA;AACJ,CAAC;AAhBD,sCAgBC;AAED,6BACE,SAAoB,EACpB,MAA0B;;;;;;oBAEpB,YAAY,GAAG,MAAM,IAAK,QAAQ,CAAC,aAAa,CAAC,QAAQ,CAAC,CAAA;oBAE1D,KAA+B,SAAS,CAAC,KAAK,CAAC,KAAK,CAAC,qBAAU,CAAC,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAlF,MAAM,QAAA,EAAE,KAAK,QAAA,EAAE,WAAW,QAAA,CAAwD;oBACzF,qBAAM,EAAE,CAAC,QAAQ,CAAC,SAAS,CAAC,IAAI,CAAC,MAAM,EAAE,KAAK,EAAE,WAAW,CAAC,CAAC,KAAK,EAAE,EAAE,YAAY,CAAC,EAAA;;oBAAnF,SAAmF,CAAA;oBAEnF,sBAAO,YAAY,EAAA;;;;CACpB;AAVD,kDAUC"}
\ No newline at end of file

examples/public/commons.js

@@ -113,6 +113,10 @@ function renderNavBar(navbarId, exampleUri) {
     {
       uri: 'detect_and_recognize_faces',
       name: 'Detect and Recognize Faces'
+    },
+    {
+      uri: 'batch_face_landmarks',
+      name: 'Batch Face Landmarks'
     }
   ]
 
@@ -152,9 +156,11 @@ function renderNavBar(navbarId, exampleUri) {
   menuContent.appendChild(li)
 
   examples
-    .filter(ex => ex.uri !== exampleUri)
     .forEach(ex => {
       const li = document.createElement('li')
+      if (ex.uri === exampleUri) {
+        li.style.background='#b0b0b0'
+      }
       const a = document.createElement('a')
       a.classList.add('waves-effect', 'waves-light')
       a.href = ex.uri

examples/server.js

@@ -24,6 +24,9 @@ app.get('/detect_and_draw_faces', (req, res) => res.sendFile(path.join(viewsDir,
 app.get('/detect_and_draw_landmarks', (req, res) => res.sendFile(path.join(viewsDir, 'detectAndDrawLandmarks.html')))
 app.get('/face_alignment', (req, res) => res.sendFile(path.join(viewsDir, 'faceAlignment.html')))
 app.get('/detect_and_recognize_faces', (req, res) => res.sendFile(path.join(viewsDir, 'detectAndRecognizeFaces.html')))
+app.get('/batch_face_landmarks', (req, res) => res.sendFile(path.join(viewsDir, 'batchFaceLandmarks.html')))
+
+
 
 app.post('/fetch_external_image', async (req, res) => {
   const { imageUrl } = req.body

examples/views/batchFaceLandmarks.html

+<!DOCTYPE html>
+<html>
+<head>
+  <script src="face-api.js"></script>
+  <script src="commons.js"></script>
+  <link rel="stylesheet" href="styles.css">
+  <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/materialize/0.100.2/css/materialize.css">
+  <script type="text/javascript" src="https://code.jquery.com/jquery-2.1.1.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/materialize/0.100.2/js/materialize.min.js"></script>
+</head>
+<body>
+  <div id="navbar"></div>
+  <div class="center-content page-container">
+    <div>
+      <div class="progress" id="loader">
+        <div class="indeterminate"></div>
+      </div>
+      <div class="row side-by-side">
+        <div class="row">
+          <label for="timeNoBatch">Time for processing each face seperately:</label>
+          <input disabled value="-" id="timeNoBatch" type="text" class="bold"/>
+        </div>
+        <div class="row">
+          <label for="timeBatch">Time for processing in Batch:</label>
+          <input disabled value="-" id="timeBatch" type="text" class="bold"/>
+        </div>
+      </div>
+      <div class="row side-by-side">
+        <div>
+          <label for="numImages">Num Images:</label>
+          <input id="numImages" type="text" class="bold" value="40"/>
+        </div>
+        <button
+          class="waves-effect waves-light btn"
+          onclick="measureTimingsAndDisplay()"
+        >
+          Ok
+        </button>
+      </div>
+      <div class="row side-by-side">
+        <div class="center-content">
+          <div id="faceContainer"></div>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <script>
+    let images = []
+    let landmarksByFace = []
+    let numImages = 40
+
+    function onNumImagesChanged(e) {
+      const val = parseInt(e.target.value) || 40
+      numImages = Math.min(Math.max(val, 0), 40)
+      e.target.value = numImages
+    }
+
+    function displayTimeStats(timeNoBatch, timeBatch) {
+      $('#timeNoBatch').val(`${timeNoBatch} ms`)
+      $('#timeBatch').val(`${timeBatch} ms`)
+    }
+
+    function drawLandmarkCanvas(img, landmarks) {
+      const canvas = faceapi.createCanvasFromMedia(img)
+      $('#faceContainer').append(canvas)
+      faceapi.drawLandmarks(canvas, landmarks, { lineWidth: 2 , drawLines: true })
+    }
+
+    async function runLandmarkDetection(useBatchInput) {
+      const ts = Date.now()
+      landmarksByFace = useBatchInput
+        ? await faceapi.detectLandmarks(images.slice(0, numImages))
+        : await Promise.all(images.slice(0, numImages).map(img => faceapi.detectLandmarks(img)))
+      const time = Date.now() - ts
+      return time
+    }
+
+    async function measureTimings() {
+      const timeNoBatch = await runLandmarkDetection(false)
+      const timeBatch = await runLandmarkDetection(true)
+      return { timeNoBatch, timeBatch }
+    }
+
+    async function measureTimingsAndDisplay() {
+      const { timeNoBatch, timeBatch } = await measureTimings()
+      displayTimeStats(timeNoBatch, timeBatch)
+      $('#faceContainer').empty()
+      landmarksByFace.forEach((landmarks, i) => drawLandmarkCanvas(images[i], landmarks))
+    }
+
+    async function run() {
+      await faceapi.loadFaceLandmarkModel('/')
+      $('#loader').hide()
+      const allImgUris = classes
+        .map(clazz => Array.from(Array(5), (_, idx) => getFaceImageUri(clazz, idx + 1)))
+        .reduce((flat, arr) => flat.concat(arr))
+
+      images = await Promise.all(allImgUris.map(
+        async uri => faceapi.bufferToImage(await fetchImage(uri))
+      ))
+      // warmup
+      await measureTimings()
+      // run
+      measureTimingsAndDisplay()
+    }
+
+    $(document).ready(function() {
+      $('#numImages').on('change', onNumImagesChanged)
+      renderNavBar('#navbar', 'batch_face_landmarks')
+      run()
+    })
+
+  </script>
+
+</body>
+</html>
\ No newline at end of file

examples/views/detectAndDrawFaces.html

@@ -84,9 +84,12 @@
       const detections = await faceapi.locateFaces(input, minConfidence)
       faceapi.drawDetection('overlay', detections.map(det => det.forSize(width, height)))
 
-      const faceImages = await faceapi.extractFaces(input.canvases[0], detections)
+      const faceImages = await faceapi.extractFaces(input.inputs[0], detections)
       $('#facesContainer').empty()
       faceImages.forEach(canvas => $('#facesContainer').append(canvas))
+
+      // free memory for input tensors
+      input.dispose()
     }
 
     async function onSelectionChanged(uri) {

examples/views/detectAndDrawLandmarks.html

@@ -103,6 +103,9 @@
 
       faceapi.drawLandmarks(canvas, landmarksByFace, { lineWidth: drawLines ? 2 : 4, drawLines, color: 'red' })
       faceapi.drawDetection('overlay', locations.map(det => det.forSize(width, height)))
+
+      // free memory for input tensors
+      input.dispose()
     }
 
     async function run() {

examples/views/faceAlignment.html

@@ -84,7 +84,7 @@
       const input = await faceapi.toNetInput(inputImgEl)
       const locations = await faceapi.locateFaces(input, minConfidence)
 
-      const faceImages = await faceapi.extractFaces(input.canvases[0], locations)
+      const faceImages = await faceapi.extractFaces(input.inputs[0], locations)
 
       // detect landmarks and get the aligned face image bounding boxes
       const alignedFaceBoxes = await Promise.all(faceImages.map(
@@ -93,7 +93,10 @@
           return faceLandmarks.align(locations[i])
         }
       ))
-      const alignedFaceImages = await faceapi.extractFaces(input.canvases[0], alignedFaceBoxes)
+      const alignedFaceImages = await faceapi.extractFaces(input.inputs[0], alignedFaceBoxes)
+
+      // free memory for input tensors
+      input.dispose()
 
       $('#facesContainer').empty()
       faceImages.forEach(async (faceCanvas, i) => {

examples/views/faceDetectionVideo.html

@@ -72,14 +72,13 @@
       if(videoEl.paused || videoEl.ended || !modelLoaded)
         return false
 
-      const input = await faceapi.toNetInput(videoEl)
-      const { width, height } = input
+      const { width, height } = faceapi.getMediaDimensions(videoEl)
       const canvas = $('#overlay').get(0)
       canvas.width = width
       canvas.height = height
 
       const ts = Date.now()
-      result = await faceapi.locateFaces(input, minConfidence)
+      result = await faceapi.locateFaces(videoEl, minConfidence)
       displayTimeStats(Date.now() - ts)
 
       faceapi.drawDetection('overlay', result.map(det => det.forSize(width, height)))

karma.conf.js

@@ -2,6 +2,7 @@ const dataFiles = [
   'test/images/*.jpg',
   'test/images/*.png',
   'test/data/*.json',
+  'test/media/*.mp4',
   'weights/**/*'
 ].map(pattern => ({
   pattern,
@@ -24,6 +25,12 @@ module.exports = function(config) {
     karmaTypescriptConfig: {
       tsconfig: 'tsconfig.test.json'
     },
-    browsers: ['Chrome']
+    browsers: ['Chrome'],
+    browserNoActivityTimeout: 60000,
+    client: {
+      jasmine: {
+        timeoutInterval: 30000
+      }
+    }
   })
 }

src/NetInput.ts

-import { Dimensions, TMediaElement } from './types';
+import * as tf from '@tensorflow/tfjs-core';
+
+import { isTensor3D, isTensor4D } from './commons/isTensor';
+import { padToSquare } from './padToSquare';
+import { Point } from './Point';
+import { TResolvedNetInput } from './types';
 import { createCanvasFromMedia } from './utils';
 
 export class NetInput {
-  private _canvases: HTMLCanvasElement[]
+  private _inputs: tf.Tensor3D[] = []
+  private _isManaged: boolean = false
+  private _isBatchInput: boolean = false
+
+  private _inputDimensions: number[][] = []
+  private _paddings: Point[] = []
 
   constructor(
-    medias: Array<TMediaElement>,
-    dims?: Dimensions
+    inputs: tf.Tensor4D | Array<TResolvedNetInput>,
+    isBatchInput: boolean = false
   ) {
-    this._canvases = []
-    medias.forEach(m => this.initCanvas(m, dims))
-  }
+    if (isTensor4D(inputs)) {
+      this._inputs = tf.unstack(inputs as tf.Tensor4D) as tf.Tensor3D[]
+    }
 
-  private initCanvas(media: TMediaElement, dims?: Dimensions) {
-    if (media instanceof HTMLCanvasElement) {
-      this._canvases.push(media)
-      return
+    if (Array.isArray(inputs)) {
+      this._inputs = inputs.map(input => {
+        if (isTensor3D(input)) {
+          // TODO: make sure not to dispose original tensors passed in by the user
+          return tf.clone(input as tf.Tensor3D)
+        }
+
+        return tf.fromPixels(
+          input instanceof HTMLCanvasElement ? input : createCanvasFromMedia(input as HTMLImageElement | HTMLVideoElement)
+        )
+      })
     }
 
-    // if input is batch type, make sure every canvas has the same dimensions
-    const canvasDims = this.dims || dims
-    this._canvases.push(createCanvasFromMedia(media, canvasDims))
+    this._isBatchInput = this.batchSize > 1 || isBatchInput
+    this._inputDimensions = this._inputs.map(t => t.shape)
+  }
+
+  public get inputs(): tf.Tensor3D[] {
+    return this._inputs
+  }
+
+  public get isManaged(): boolean {
+    return this._isManaged
+  }
+
+  public get isBatchInput(): boolean {
+    return this._isBatchInput
+  }
+
+  public get batchSize(): number {
+    return this._inputs.length
+  }
+
+  public get inputDimensions(): number[][] {
+    return this._inputDimensions
+  }
+
+  public get paddings(): Point[] {
+    return this._paddings
+  }
+
+  public getInputDimensions(batchIdx: number): number[] {
+    return this._inputDimensions[batchIdx]
+  }
+
+  public getInputHeight(batchIdx: number): number {
+    return this._inputDimensions[batchIdx][0]
+  }
+
+  public getInputWidth(batchIdx: number): number {
+    return this._inputDimensions[batchIdx][1]
   }
 
-  public get canvases() : HTMLCanvasElement[] {
-    return this._canvases
+  public getPaddings(batchIdx: number): Point {
+    return this._paddings[batchIdx]
   }
 
-  public get width() : number {
-    return (this._canvases[0] || {}).width
+  public toBatchTensor(inputSize: number, isCenterInputs: boolean = true): tf.Tensor4D {
+
+    return tf.tidy(() => {
+
+      const inputTensors = this._inputs.map((inputTensor: tf.Tensor3D) => {
+        const [originalHeight, originalWidth] = inputTensor.shape
+
+        let imgTensor = inputTensor.expandDims().toFloat() as tf.Tensor4D
+        imgTensor = padToSquare(imgTensor, isCenterInputs)
+
+        const [heightAfterPadding, widthAfterPadding] = imgTensor.shape.slice(1)
+
+        if (heightAfterPadding !== inputSize || widthAfterPadding !== inputSize) {
+          imgTensor = tf.image.resizeBilinear(imgTensor, [inputSize, inputSize])
+        }
+
+        this._paddings.push(new Point(
+          widthAfterPadding - originalWidth,
+          heightAfterPadding - originalHeight
+        ))
+        return imgTensor
+      })
+
+      const batchTensor = tf.stack(inputTensors).as4D(this.batchSize, inputSize, inputSize, 3)
+
+      if (this.isManaged) {
+        this.dispose()
+      }
+
+      return batchTensor
+    })
   }
 
-  public get height() : number {
-    return (this._canvases[0] || {}).height
+  /**
+   *  By setting the isManaged flag, all newly created tensors will be automatically
+   *  automatically disposed after the batch tensor has been created
+   */
+  public managed() {
+    this._isManaged = true
+    return this
   }
 
-  public get dims() : Dimensions | null {
-    const { width, height } = this
-    return (width > 0 && height > 0) ? { width, height } : null
+  public dispose() {
+    this._inputs.forEach(t => t.dispose())
   }
 }
\ No newline at end of file

src/allFacesFactory.ts

-import * as tf from '@tensorflow/tfjs-core';
-
 import { extractFaceTensors } from './extractFaceTensors';
 import { FaceDetectionNet } from './faceDetectionNet/FaceDetectionNet';
 import { FaceLandmarkNet } from './faceLandmarkNet/FaceLandmarkNet';
+import { FaceLandmarks } from './faceLandmarkNet/FaceLandmarks';
 import { FaceRecognitionNet } from './faceRecognitionNet/FaceRecognitionNet';
 import { FullFaceDescription } from './FullFaceDescription';
-import { NetInput } from './NetInput';
 import { TNetInput } from './types';
 
 export function allFacesFactory(
@@ -14,21 +12,25 @@ export function allFacesFactory(
   recognitionNet: FaceRecognitionNet
 ) {
   return async function(
-    input: tf.Tensor | NetInput | TNetInput,
+    input: TNetInput,
     minConfidence: number
   ): Promise<FullFaceDescription[]> {
 
     const detections = await detectionNet.locateFaces(input, minConfidence)
 
     const faceTensors = await extractFaceTensors(input, detections)
+    /**
     const faceLandmarksByFace = await Promise.all(faceTensors.map(
       faceTensor => landmarkNet.detectLandmarks(faceTensor)
-    ))
+    )) as FaceLandmarks[]
+     */
+    const faceLandmarksByFace = await landmarkNet.detectLandmarks(faceTensors) as FaceLandmarks[]
+
     faceTensors.forEach(t => t.dispose())
 
-    const alignedFaceBoxes = await Promise.all(faceLandmarksByFace.map(
+    const alignedFaceBoxes = faceLandmarksByFace.map(
       (landmarks, i) => landmarks.align(detections[i].getBox())
-    ))
+    )
     const alignedFaceTensors = await extractFaceTensors(input, alignedFaceBoxes)
 
     const descriptors = await Promise.all(alignedFaceTensors.map(

src/commons/getImageTensor.ts

-import * as tf from '@tensorflow/tfjs-core';
-
-import { NetInput } from '../NetInput';
-
-export function getImageTensor(input: tf.Tensor | NetInput): tf.Tensor4D {
-  return tf.tidy(() => {
-    if (input instanceof tf.Tensor) {
-      const rank = input.shape.length
-      if (rank !== 3 && rank !== 4) {
-        throw new Error('input tensor must be of rank 3 or 4')
-      }
-
-      return (rank === 3 ? input.expandDims(0) : input).toFloat() as tf.Tensor4D
-    }
-
-    if (!(input instanceof NetInput)) {
-      throw new Error('getImageTensor - expected input to be a tensor or an instance of NetInput')
-    }
-
-    return tf.concat(
-      input.canvases.map(canvas =>
-        tf.fromPixels(canvas).expandDims(0).toFloat()
-      )
-    ) as tf.Tensor4D
-  })
-}
\ No newline at end of file

src/commons/isMediaElement.ts

+export function isMediaElement(input: any) {
+  return input instanceof HTMLImageElement
+    || input instanceof HTMLVideoElement
+    || input instanceof HTMLCanvasElement
+}
\ No newline at end of file

src/commons/isTensor.ts

 import * as tf from '@tensorflow/tfjs-core';
 
-export function isTensor(tensor: tf.Tensor, dim: number) {
+export function isTensor(tensor: any, dim: number) {
   return tensor instanceof tf.Tensor && tensor.shape.length === dim
 }
 
-export function isTensor1D(tensor: tf.Tensor) {
+export function isTensor1D(tensor: any) {
   return isTensor(tensor, 1)
 }
 
-export function isTensor2D(tensor: tf.Tensor) {
+export function isTensor2D(tensor: any) {
   return isTensor(tensor, 2)
 }
 
-export function isTensor3D(tensor: tf.Tensor) {
+export function isTensor3D(tensor: any) {
   return isTensor(tensor, 3)
 }
 
-export function isTensor4D(tensor: tf.Tensor) {
+export function isTensor4D(tensor: any) {
   return isTensor(tensor, 4)
 }
\ No newline at end of file

src/commons/types.ts

@@ -5,4 +5,11 @@ export type ConvParams = {
   bias: tf.Tensor1D
 }
 
-export type ExtractWeightsFunction = (numWeights: number) => Float32Array
\ No newline at end of file
+export type ExtractWeightsFunction = (numWeights: number) => Float32Array
+
+export type BatchReshapeInfo = {
+  originalWidth: number
+  originalHeight: number
+  paddingX: number
+  paddingY: number
+}

src/drawing/index.ts

 import { FaceDetection } from '../faceDetectionNet/FaceDetection';
 import { FaceLandmarks } from '../faceLandmarkNet/FaceLandmarks';
 import { Point } from '../Point';
-import { getContext2dOrThrow, getElement, round } from '../utils';
+import { getContext2dOrThrow, resolveInput, round } from '../utils';
 import { DrawBoxOptions, DrawLandmarksOptions, DrawOptions, DrawTextOptions } from './types';
 
 export function getDefaultDrawOptions(): DrawOptions {
@@ -55,7 +55,7 @@ export function drawDetection(
   detection: FaceDetection | FaceDetection[],
   options?: DrawBoxOptions & DrawTextOptions & { withScore: boolean }
 ) {
-  const canvas = getElement(canvasArg)
+  const canvas = resolveInput(canvasArg)
   if (!(canvas instanceof HTMLCanvasElement)) {
     throw new Error('drawBox - expected canvas to be of type: HTMLCanvasElement')
   }
@@ -132,7 +132,7 @@ export function drawLandmarks(
   faceLandmarks: FaceLandmarks | FaceLandmarks[],
   options?: DrawLandmarksOptions & { drawLines: boolean }
 ) {
-  const canvas = getElement(canvasArg)
+  const canvas = resolveInput(canvasArg)
   if (!(canvas instanceof HTMLCanvasElement)) {
     throw new Error('drawLandmarks - expected canvas to be of type: HTMLCanvasElement')
   }

src/extractFaceTensors.ts

 import * as tf from '@tensorflow/tfjs-core';
 
-import { getImageTensor } from './commons/getImageTensor';
 import { FaceDetection } from './faceDetectionNet/FaceDetection';
-import { NetInput } from './NetInput';
 import { Rect } from './Rect';
-import { TNetInput } from './types';
 import { toNetInput } from './toNetInput';
+import { TNetInput } from './types';
 
 /**
  * Extracts the tensors of the image regions containing the detected faces.
@@ -18,19 +16,23 @@ import { toNetInput } from './toNetInput';
  * @returns Tensors of the corresponding image region for each detected face.
  */
 export async function extractFaceTensors(
-  input: tf.Tensor | NetInput | TNetInput,
-  detections: Array<FaceDetection|Rect>
+  input: TNetInput,
+  detections: Array<FaceDetection | Rect>
 ): Promise<tf.Tensor4D[]> {
 
-  const image = input instanceof tf.Tensor
-    ? input
-    : await toNetInput(input)
+  const netInput = await toNetInput(input, true)
+
+  if (netInput.batchSize > 1) {
+    if (netInput.isManaged) {
+      netInput.dispose()
+    }
+    throw new Error('extractFaceTensors - batchSize > 1 not supported')
+  }
 
   return tf.tidy(() => {
-    const imgTensor = getImageTensor(image)
+    const imgTensor = netInput.inputs[0].expandDims().toFloat() as tf.Tensor4D
 
-    // TODO handle batches
-    const [batchSize, imgHeight, imgWidth, numChannels] = imgTensor.shape
+    const [imgHeight, imgWidth, numChannels] = imgTensor.shape.slice(1)
 
     const boxes = detections.map(
       det => det instanceof FaceDetection
@@ -41,6 +43,9 @@ export async function extractFaceTensors(
       tf.slice(imgTensor, [0, y, x, 0], [1, height, width, numChannels])
     )
 
+    if (netInput.isManaged) {
+      netInput.dispose()
+    }
     return faceTensors
   })
 }
\ No newline at end of file

src/extractFaces.ts

 import { FaceDetection } from './faceDetectionNet/FaceDetection';
 import { Rect } from './Rect';
-import { createCanvas, getContext2dOrThrow } from './utils';
+import { toNetInput } from './toNetInput';
+import { TNetInput } from './types';
+import { createCanvas, getContext2dOrThrow, imageTensorToCanvas } from './utils';
 
 /**
  * Extracts the image regions containing the detected faces.
@@ -9,15 +11,31 @@ import { createCanvas, getContext2dOrThrow } from './utils';
  * @param detections The face detection results or face bounding boxes for that image.
  * @returns The Canvases of the corresponding image region for each detected face.
  */
-export function extractFaces(
-  image: HTMLCanvasElement,
-  detections: Array<FaceDetection|Rect>
-): HTMLCanvasElement[] {
-  const ctx = getContext2dOrThrow(image)
+export async function extractFaces(
+  input: TNetInput,
+  detections: Array<FaceDetection | Rect>
+): Promise<HTMLCanvasElement[]> {
+
+  let canvas = input as HTMLCanvasElement
+
+  if (!(input instanceof HTMLCanvasElement)) {
+    const netInput = await toNetInput(input, true)
+
+    if (netInput.batchSize > 1) {
+      if (netInput.isManaged) {
+        netInput.dispose()
+      }
+      throw new Error('extractFaces - batchSize > 1 not supported')
+    }
+
+    canvas = await imageTensorToCanvas(netInput.inputs[0])
+  }
+
+  const ctx = getContext2dOrThrow(canvas)
 
   const boxes = detections.map(
     det => det instanceof FaceDetection
-      ? det.forSize(image.width, image.height).getBox().floor()
+      ? det.forSize(canvas.width, canvas.height).getBox().floor()
       : det
   )
   return boxes.map(({ x, y, width, height }) => {

src/faceDetectionNet/FaceDetectionNet.ts

 import * as tf from '@tensorflow/tfjs-core';
 
-import { getImageTensor } from '../commons/getImageTensor';
 import { NetInput } from '../NetInput';
-import { padToSquare } from '../padToSquare';
 import { Rect } from '../Rect';
 import { toNetInput } from '../toNetInput';
-import { Dimensions, TNetInput } from '../types';
+import { TNetInput } from '../types';
 import { extractParams } from './extractParams';
 import { FaceDetection } from './FaceDetection';
 import { loadQuantizedParams } from './loadQuantizedParams';
@@ -13,7 +11,6 @@ import { mobileNetV1 } from './mobileNetV1';
 import { nonMaxSuppression } from './nonMaxSuppression';
 import { outputLayer } from './outputLayer';
 import { predictionLayer } from './predictionLayer';
-import { resizeLayer } from './resizeLayer';
 import { NetParams } from './types';
 
 export class FaceDetectionNet {
@@ -36,15 +33,16 @@ export class FaceDetectionNet {
     this._params = extractParams(weights)
   }
 
-  private forwardTensor(imgTensor: tf.Tensor4D) {
+  public forwardInput(input: NetInput) {
     if (!this._params) {
       throw new Error('FaceDetectionNet - load model before inference')
     }
 
     return tf.tidy(() => {
+      const batchTensor = input.toBatchTensor(512, false)
 
-      const resized = resizeLayer(imgTensor) as tf.Tensor4D
-      const features = mobileNetV1(resized, this._params.mobilenetv1_params)
+      const x = tf.sub(tf.mul(batchTensor, tf.scalar(0.007843137718737125)), tf.scalar(1)) as tf.Tensor4D
+      const features = mobileNetV1(x, this._params.mobilenetv1_params)
 
       const {
         boxPredictions,
@@ -55,44 +53,23 @@ export class FaceDetectionNet {
     })
   }
 
-  public async forward(input: tf.Tensor | NetInput | TNetInput) {
-    const netInput = input instanceof tf.Tensor
-      ? input
-      : await toNetInput(input)
-
-    return tf.tidy(() =>
-      this.forwardTensor(padToSquare(getImageTensor(netInput)))
-    )
+  public async forward(input: TNetInput) {
+    return this.forwardInput(await toNetInput(input, true))
   }
 
   public async locateFaces(
-    input: tf.Tensor | NetInput | TNetInput,
+    input: TNetInput,
     minConfidence: number = 0.8,
     maxResults: number = 100,
   ): Promise<FaceDetection[]> {
 
-    const netInput = input instanceof tf.Tensor
-      ? input
-      : await toNetInput(input)
-
-    let paddedHeightRelative = 1, paddedWidthRelative = 1
-    let imageDimensions: Dimensions | undefined
+    const netInput = await toNetInput(input, true)
 
     const {
       boxes: _boxes,
       scores: _scores
-    } = tf.tidy(() => {
-
-      let imgTensor = getImageTensor(netInput)
-      const [height, width] = imgTensor.shape.slice(1)
-      imageDimensions = { width, height }
+    } = this.forwardInput(netInput)
 
-      imgTensor = padToSquare(imgTensor)
-      paddedHeightRelative = imgTensor.shape[1] / height
-      paddedWidthRelative = imgTensor.shape[2] / width
-
-      return this.forwardTensor(imgTensor)
-    })
 
     // TODO batches
     const boxes = _boxes[0]
@@ -114,6 +91,10 @@ export class FaceDetectionNet {
       minConfidence
     )
 
+
+    const paddedHeightRelative = (netInput.getPaddings(0).y + netInput.getInputHeight(0)) / netInput.getInputHeight(0)
+    const paddedWidthRelative = (netInput.getPaddings(0).x + netInput.getInputWidth(0)) / netInput.getInputWidth(0)
+
     const results = indices
       .map(idx => {
         const [top, bottom] = [
@@ -132,7 +113,10 @@ export class FaceDetectionNet {
             right - left,
             bottom - top
           ),
-          imageDimensions as Dimensions
+          {
+            height: netInput.getInputHeight(0),
+            width: netInput.getInputWidth(0)
+          }
         )
       })
 

src/faceDetectionNet/resizeLayer.ts

-import * as tf from '@tensorflow/tfjs-core';
-
-const resizedImageSize = [512, 512] as [number, number]
-const weight = tf.scalar(0.007843137718737125)
-const bias = tf.scalar(1)
-
-export function resizeLayer(x: tf.Tensor4D) {
-  return tf.tidy(() => {
-
-    const resized = tf.image.resizeBilinear(x, resizedImageSize, false)
-    return tf.sub(tf.mul(resized, weight), bias)
-
-  })
-}
\ No newline at end of file

src/faceLandmarkNet/FaceLandmarkNet.ts

 import * as tf from '@tensorflow/tfjs-core';
 
 import { convLayer } from '../commons/convLayer';
-import { getImageTensor } from '../commons/getImageTensor';
 import { ConvParams } from '../commons/types';
 import { NetInput } from '../NetInput';
-import { padToSquare } from '../padToSquare';
 import { Point } from '../Point';
 import { toNetInput } from '../toNetInput';
-import { Dimensions, TNetInput } from '../types';
+import { TNetInput } from '../types';
 import { isEven } from '../utils';
 import { extractParams } from './extractParams';
 import { FaceLandmarks } from './FaceLandmarks';
@@ -43,7 +41,7 @@ export class FaceLandmarkNet {
     this._params = extractParams(weights)
   }
 
-  public forwardTensor(imgTensor: tf.Tensor4D): tf.Tensor2D {
+  public forwardInput(input: NetInput): tf.Tensor2D {
     const params = this._params
 
     if (!params) {
@@ -51,17 +49,9 @@ export class FaceLandmarkNet {
     }
 
     return tf.tidy(() => {
-      const [batchSize, height, width] = imgTensor.shape.slice()
+      const batchTensor = input.toBatchTensor(128, true)
 
-      let x = padToSquare(imgTensor, true)
-      const [heightAfterPadding, widthAfterPadding] = x.shape.slice(1)
-
-      // work with 128 x 128 sized face images
-      if (heightAfterPadding !== 128 || widthAfterPadding !== 128) {
-        x = tf.image.resizeBilinear(x, [128, 128])
-      }
-
-      let out = conv(x, params.conv0_params)
+      let out = conv(batchTensor, params.conv0_params)
       out = maxPool(out)
       out = conv(out, params.conv1_params)
       out = conv(out, params.conv2_params)
@@ -76,62 +66,72 @@ export class FaceLandmarkNet {
       const fc0 = tf.relu(fullyConnectedLayer(out.as2D(out.shape[0], -1), params.fc0_params))
       const fc1 = fullyConnectedLayer(fc0, params.fc1_params)
 
-
       const createInterleavedTensor = (fillX: number, fillY: number) =>
-      tf.stack([
-        tf.fill([68], fillX),
-        tf.fill([68], fillY)
-      ], 1).as2D(batchSize, 136)
-
+        tf.stack([
+          tf.fill([68], fillX),
+          tf.fill([68], fillY)
+        ], 1).as2D(1, 136).as1D()
 
       /* shift coordinates back, to undo centered padding
-        ((x * widthAfterPadding) - shiftX) / width
-        ((y * heightAfterPadding) - shiftY) / height
+        x = ((x * widthAfterPadding) - shiftX) / width
+        y = ((y * heightAfterPadding) - shiftY) / height
       */
-      const shiftX = Math.floor(Math.abs(widthAfterPadding - width) / 2)
-      const shiftY = Math.floor(Math.abs(heightAfterPadding - height) / 2)
-      const landmarkTensor = fc1
-        .mul(createInterleavedTensor(widthAfterPadding, heightAfterPadding))
-        .sub(createInterleavedTensor(shiftX, shiftY))
-        .div(createInterleavedTensor(width, height))
-
-      return landmarkTensor as tf.Tensor2D
+
+      const landmarkTensors = fc1
+        .mul(tf.stack(Array.from(Array(input.batchSize), (_, batchIdx) =>
+          createInterleavedTensor(
+            input.getPaddings(batchIdx).x + input.getInputWidth(batchIdx),
+            input.getPaddings(batchIdx).y + input.getInputHeight(batchIdx)
+          )
+        )))
+        .sub(tf.stack(Array.from(Array(input.batchSize), (_, batchIdx) =>
+          createInterleavedTensor(
+            Math.floor(input.getPaddings(batchIdx).x / 2),
+            Math.floor(input.getPaddings(batchIdx).y / 2)
+          )
+        )))
+        .div(tf.stack(Array.from(Array(input.batchSize), (_, batchIdx) =>
+          createInterleavedTensor(
+            input.getInputWidth(batchIdx),
+            input.getInputHeight(batchIdx)
+          )
+        )))
+
+      return landmarkTensors as tf.Tensor2D
     })
   }
 
-  public async forward(input: tf.Tensor | NetInput | TNetInput): Promise<tf.Tensor2D> {
-    const netInput = input instanceof tf.Tensor
-      ? input
-      : await toNetInput(input)
-
-    return this.forwardTensor(getImageTensor(netInput))
+  public async forward(input: TNetInput): Promise<tf.Tensor2D> {
+    return this.forwardInput(await toNetInput(input, true))
   }
 
-  public async detectLandmarks(input: tf.Tensor | NetInput | TNetInput) {
-    const netInput = input instanceof tf.Tensor
-      ? input
-      : await toNetInput(input)
+  public async detectLandmarks(input: TNetInput): Promise<FaceLandmarks | FaceLandmarks[]> {
+    const netInput = await toNetInput(input, true)
 
-    let imageDimensions: Dimensions | undefined
-
-    const outTensor = tf.tidy(() => {
-      const imgTensor = getImageTensor(netInput)
-
-      const [height, width] = imgTensor.shape.slice(1)
-      imageDimensions = { width, height }
-
-      return this.forwardTensor(imgTensor)
-    })
+    const landmarkTensors = tf.tidy(
+      () => tf.unstack(this.forwardInput(netInput))
+    )
 
-    const faceLandmarksArray = Array.from(await outTensor.data())
-    outTensor.dispose()
+    const landmarksForBatch = await Promise.all(landmarkTensors.map(
+      async (landmarkTensor, batchIdx) => {
+        const landmarksArray = Array.from(await landmarkTensor.data())
+        const xCoords = landmarksArray.filter((_, i) => isEven(i))
+        const yCoords = landmarksArray.filter((_, i) => !isEven(i))
+
+        return new FaceLandmarks(
+          Array(68).fill(0).map((_, i) => new Point(xCoords[i], yCoords[i])),
+          {
+            height: netInput.getInputHeight(batchIdx),
+            width : netInput.getInputWidth(batchIdx),
+          }
+        )
+      }
+    ))
 
-    const xCoords = faceLandmarksArray.filter((_, i) => isEven(i))
-    const yCoords = faceLandmarksArray.filter((_, i) => !isEven(i))
+    landmarkTensors.forEach(t => t.dispose())
 
-    return new FaceLandmarks(
-      Array(68).fill(0).map((_, i) => new Point(xCoords[i], yCoords[i])),
-      imageDimensions as Dimensions
-    )
+    return netInput.isBatchInput
+      ? landmarksForBatch
+      : landmarksForBatch[0]
   }
 }
\ No newline at end of file

src/faceRecognitionNet/FaceRecognitionNet.ts

 import * as tf from '@tensorflow/tfjs-core';
 
-import { getImageTensor } from '../commons/getImageTensor';
 import { NetInput } from '../NetInput';
-import { padToSquare } from '../padToSquare';
 import { toNetInput } from '../toNetInput';
 import { TNetInput } from '../types';
 import { convDown } from './convLayer';
@@ -32,25 +30,18 @@ export class FaceRecognitionNet {
     this._params = extractParams(weights)
   }
 
-  public async forward(input: tf.Tensor | NetInput | TNetInput): Promise<tf.Tensor2D> {
+  public async forwardInput(input: NetInput): Promise<tf.Tensor2D> {
     if (!this._params) {
       throw new Error('FaceRecognitionNet - load model before inference')
     }
 
-    const netInput = input instanceof tf.Tensor
-      ? input
-      : await toNetInput(input)
 
     return tf.tidy(() => {
+      const batchTensor = input.toBatchTensor(150, true)
 
-      let x = padToSquare(getImageTensor(netInput), true)
-      // work with 150 x 150 sized face images
-      if (x.shape[1] !== 150 || x.shape[2] !== 150) {
-        x = tf.image.resizeBilinear(x, [150, 150])
-      }
-      x = normalize(x)
+      const normalized = normalize(batchTensor)
 
-      let out = convDown(x, this._params.conv32_down)
+      let out = convDown(normalized, this._params.conv32_down)
       out = tf.maxPool(out, 3, 2, 'valid')
 
       out = residual(out, this._params.conv32_1)
@@ -77,13 +68,12 @@ export class FaceRecognitionNet {
       return fullyConnected
     })
   }
+  public async forward(input: TNetInput): Promise<tf.Tensor2D> {
+    return this.forwardInput(await toNetInput(input, true))
+  }
 
-  public async computeFaceDescriptor(input: tf.Tensor | NetInput | TNetInput) {
-    const netInput = input instanceof tf.Tensor
-      ? input
-      : await toNetInput(input)
-
-    const result = await this.forward(netInput)
+  public async computeFaceDescriptor(input: TNetInput) {
+    const result = await this.forward(await toNetInput(input, true))
     const data = await result.data()
     result.dispose()
     return data as Float32Array

src/globalApi.ts

@@ -6,9 +6,9 @@ import { FaceDetectionNet } from './faceDetectionNet/FaceDetectionNet';
 import { FaceLandmarkNet } from './faceLandmarkNet/FaceLandmarkNet';
 import { FaceLandmarks } from './faceLandmarkNet/FaceLandmarks';
 import { FaceRecognitionNet } from './faceRecognitionNet/FaceRecognitionNet';
+import { FullFaceDescription } from './FullFaceDescription';
 import { NetInput } from './NetInput';
 import { TNetInput } from './types';
-import { FullFaceDescription } from './FullFaceDescription';
 
 export const detectionNet = new FaceDetectionNet()
 export const landmarkNet = new FaceLandmarkNet()
@@ -35,7 +35,7 @@ export function loadModels(url: string) {
 }
 
 export function locateFaces(
-  input: tf.Tensor | NetInput | TNetInput,
+  input: TNetInput,
   minConfidence?: number,
   maxResults?: number
 ): Promise<FaceDetection[]> {
@@ -43,13 +43,13 @@ export function locateFaces(
 }
 
 export function detectLandmarks(
-  input: tf.Tensor | NetInput | TNetInput
-): Promise<FaceLandmarks> {
+  input: TNetInput
+): Promise<FaceLandmarks | FaceLandmarks[]> {
   return landmarkNet.detectLandmarks(input)
 }
 
 export function computeFaceDescriptor(
-  input: tf.Tensor | NetInput | TNetInput
+  input: TNetInput
 ): Promise<Float32Array>  {
   return recognitionNet.computeFaceDescriptor(input)
 }

src/padToSquare.ts

 import * as tf from '@tensorflow/tfjs-core';
 
-import { isEven } from './utils';
-
 /**
  * Pads the smaller dimension of an image tensor with zeros, such that width === height.
  *

src/toNetInput.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { isMediaElement } from './commons/isMediaElement';
+import { isTensor3D, isTensor4D } from './commons/isTensor';
 import { NetInput } from './NetInput';
 import { TNetInput } from './types';
-import { awaitMediaLoaded, getElement } from './utils';
+import { awaitMediaLoaded, resolveInput } from './utils';
 
 /**
  * Validates the input to make sure, they are valid net inputs and awaits all media elements
  * to be finished loading.
  *
  * @param input The input, which can be a media element or an array of different media elements.
+ * @param manageCreatedInput If a new NetInput instance is created from the inputs, this flag
+ * determines, whether to set the NetInput as managed or not.
  * @returns A NetInput instance, which can be passed into one of the neural networks.
  */
 export async function toNetInput(
-  input: NetInput | TNetInput
+  inputs: TNetInput,
+  manageCreatedInput: boolean = false
 ): Promise<NetInput> {
-  if (input instanceof NetInput) {
-    return input
+  if (inputs instanceof NetInput) {
+    return inputs
+  }
+
+  const afterCreate = (netInput: NetInput) => manageCreatedInput
+    ? netInput.managed()
+    : netInput
+
+  if (isTensor4D(inputs)) {
+    return afterCreate(new NetInput(inputs as tf.Tensor4D))
   }
 
-  const mediaArgArray = Array.isArray(input)
-      ? input
-      : [input]
+  let inputArgArray = Array.isArray(inputs)
+      ? inputs
+      : [inputs]
 
-  if (!mediaArgArray.length) {
+  if (!inputArgArray.length) {
     throw new Error('toNetInput - empty array passed as input')
   }
 
-  const medias = mediaArgArray.map(getElement)
+  const getIdxHint = (idx: number) => Array.isArray(inputs) ? ` at input index ${idx}:` : ''
 
-  medias.forEach((media, i) => {
-    if (!(media instanceof HTMLImageElement || media instanceof HTMLVideoElement || media instanceof HTMLCanvasElement)) {
-      const idxHint = Array.isArray(input) ? ` at input index ${i}:` : ''
-      if (typeof mediaArgArray[i] === 'string') {
-        throw new Error(`toNetInput -${idxHint} string passed, but could not resolve HTMLElement for element id`)
+  const inputArray = inputArgArray
+    .map(resolveInput)
+    .map((input, i) => {
+      if (isTensor4D(input)) {
+        // if tf.Tensor4D is passed in the input array, the batch size has to be 1
+        const batchSize = input.shape[0]
+        if (batchSize !== 1) {
+          throw new Error(`toNetInput -${getIdxHint(i)} tf.Tensor4D with batchSize ${batchSize} passed, but not supported in input array`)
+        }
+        // to tf.Tensor3D
+        return input.reshape(input.shape.slice(1))
       }
-      throw new Error(`toNetInput -${idxHint} expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement, or to be an element id`)
+      return input
+    })
+
+  inputArray.forEach((input, i) => {
+    if (!isMediaElement(input) && !isTensor3D(input)) {
+
+      if (typeof inputArgArray[i] === 'string') {
+        throw new Error(`toNetInput -${getIdxHint(i)} string passed, but could not resolve HTMLElement for element id ${inputArgArray[i]}`)
+      }
+
+      throw new Error(`toNetInput -${getIdxHint(i)} expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement | tf.Tensor3D, or to be an element id`)
     }
   })
 
   // wait for all media elements being loaded
   await Promise.all(
-    medias.map(media => awaitMediaLoaded(media))
+    inputArray.map(input => isMediaElement(input) && awaitMediaLoaded(input))
   )
 
-  return new NetInput(medias)
+  return afterCreate(new NetInput(inputArray, Array.isArray(inputs)))
 }
\ No newline at end of file

src/types.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { NetInput } from './NetInput';
+
 export type TMediaElement = HTMLImageElement | HTMLVideoElement | HTMLCanvasElement
 
-export type TNetInputArg = string | TMediaElement
+export type TResolvedNetInput = TMediaElement | tf.Tensor3D | tf.Tensor4D
+
+export type TNetInputArg = string | TResolvedNetInput
 
-export type TNetInput = TNetInputArg | Array<TNetInputArg>
+export type TNetInput = TNetInputArg | Array<TNetInputArg> | NetInput | tf.Tensor4D
 
 export type Dimensions = {
   width: number

src/utils.ts

 import * as tf from '@tensorflow/tfjs-core';
 
+import { isTensor4D } from './commons/isTensor';
 import { Dimensions } from './types';
 
 export function isFloat(num: number) {
@@ -14,7 +15,7 @@ export function round(num: number) {
   return Math.floor(num * 100) / 100
 }
 
-export function getElement(arg: string | any) {
+export function resolveInput(arg: string | any) {
   if (typeof arg === 'string') {
     return document.getElementById(arg)
   }
@@ -106,12 +107,12 @@ export function bufferToImage(buf: Blob): Promise<HTMLImageElement> {
 }
 
 export async function imageTensorToCanvas(
-  imgTensor: tf.Tensor4D,
+  imgTensor: tf.Tensor,
   canvas?: HTMLCanvasElement
 ): Promise<HTMLCanvasElement> {
   const targetCanvas = canvas ||  document.createElement('canvas')
 
-  const [_, height, width, numChannels] = imgTensor.shape
+  const [height, width, numChannels] = imgTensor.shape.slice(isTensor4D(imgTensor) ? 1 : 0)
   await tf.toPixels(imgTensor.as3D(height, width, numChannels).toInt(), targetCanvas)
 
   return targetCanvas

test/data/faceLandmarkPositions2.json

+[{"x": 9.995004907250404, "y": 53.55449616909027}, {"x": 12.50796876847744, "y": 71.41348421573639}, {"x": 16.677917540073395, "y": 88.59677910804749}, {"x": 22.6475290954113, "y": 104.6014130115509}, {"x": 30.59161528944969, "y": 119.35952603816986}, {"x": 41.422560811042786, "y": 132.23226964473724}, {"x": 54.74700182676315, "y": 142.4335777759552}, {"x": 70.32481580972672, "y": 149.33189749717712}, {"x": 87.31497824192047, "y": 150.50972700119019}, {"x": 103.98584604263306, "y": 145.98273038864136}, {"x": 117.90181696414948, "y": 135.19554734230042}, {"x": 128.67935299873352, "y": 121.79077863693237}, {"x": 136.7296814918518, "y": 105.85636496543884}, {"x": 140.29521346092224, "y": 88.25878500938416}, {"x": 140.9232795238495, "y": 70.16736567020416}, {"x": 140.2374029159546, "y": 52.73242145776749}, {"x": 137.97148168087006, "y": 34.537942707538605}, {"x": 14.37721811234951, "y": 33.1049881875515}, {"x": 22.6781465113163, "y": 24.685607850551605}, {"x": 34.36600640416145, "y": 21.1758591234684}, {"x": 46.24761343002319, "y": 22.49436378479004}, {"x": 57.12086856365204, "y": 26.742971688508987}, {"x": 81.21025264263153, "y": 23.014162480831146}, {"x": 92.2086775302887, "y": 15.48520065844059}, {"x": 104.77548837661743, "y": 11.306393891572952}, {"x": 117.67798662185669, "y": 11.740228906273842}, {"x": 127.28274464607239, "y": 18.115675449371338}, {"x": 69.62742805480957, "y": 41.51403307914734}, {"x": 70.82946002483368, "y": 55.146731436252594}, {"x": 71.84555232524872, "y": 68.59723627567291}, {"x": 73.0046421289444, "y": 81.93029165267944}, {"x": 60.417647659778595, "y": 88.01697492599487}, {"x": 67.98770427703857, "y": 90.65443575382233}, {"x": 76.07284784317017, "y": 91.86699986457825}, {"x": 84.35145914554596, "y": 88.2117748260498}, {"x": 90.86072444915771, "y": 83.67109894752502}, {"x": 28.828849643468857, "y": 47.794362902641296}, {"x": 36.311765760183334, "y": 43.33548992872238}, {"x": 44.95347887277603, "y": 43.20283681154251}, {"x": 52.85406410694122, "y": 48.07424694299698}, {"x": 44.7566494345665, "y": 49.9691516160965}, {"x": 35.997654497623444, "y": 50.32083839178085}, {"x": 89.51361179351807, "y": 42.501528561115265}, {"x": 97.55686819553375, "y": 35.38782298564911}, {"x": 106.73499405384064, "y": 33.59129726886749}, {"x": 114.8474782705307, "y": 36.34611591696739}, {"x": 108.40394496917725, "y": 40.97002297639847}, {"x": 98.98389279842377, "y": 42.47862249612808}, {"x": 53.17014008760452, "y": 109.99322533607483}, {"x": 62.47727572917938, "y": 105.5664449930191}, {"x": 72.82306104898453, "y": 102.35638618469238}, {"x": 80.85319697856903, "y": 103.16510796546936}, {"x": 89.42103087902069, "y": 100.08856952190399}, {"x": 99.8135894536972, "y": 100.0435084104538}, {"x": 109.78849232196808, "y": 101.60946249961853}, {"x": 101.90783143043518, "y": 115.25328755378723}, {"x": 92.65078604221344, "y": 122.49988317489624}, {"x": 83.28675627708435, "y": 125.00075697898865}, {"x": 74.31002408266068, "y": 125.16917288303375}, {"x": 63.37641924619675, "y": 121.38420939445496}, {"x": 57.85811394453049, "y": 110.28821468353271}, {"x": 73.27612638473511, "y": 108.2253098487854}, {"x": 81.34059011936188, "y": 108.251291513443}, {"x": 89.9710088968277, "y": 105.99507093429565}, {"x": 104.85810041427612, "y": 103.1228095293045}, {"x": 90.87785482406616, "y": 112.19902038574219}, {"x": 82.05846548080444, "y": 114.52528238296509}, {"x": 73.8232746720314, "y": 114.64338898658752}]
\ No newline at end of file

test/data/faceLandmarkPositionsRect.json

+[{"x": 1.41040606983006, "y": 39.10264679789543}, {"x": 6.364097021520138, "y": 59.51536446809769}, {"x": 9.736957371234894, "y": 77.80020213127136}, {"x": 13.817122921347618, "y": 96.05930614471436}, {"x": 20.00808236002922, "y": 110.73724734783173}, {"x": 26.37610113620758, "y": 121.9759613275528}, {"x": 33.717534959316254, "y": 132.17582309246063}, {"x": 41.72355371713638, "y": 140.95428133010864}, {"x": 50.880420446395874, "y": 146.48704254627228}, {"x": 61.8572758436203, "y": 146.95429050922394}, {"x": 72.31115305423737, "y": 142.27578461170197}, {"x": 81.12127649784088, "y": 134.44229400157928}, {"x": 90.62704622745514, "y": 123.21754229068756}, {"x": 99.09188437461853, "y": 110.66853940486908}, {"x": 106.08789837360382, "y": 92.88722932338715}, {"x": 110.42981934547424, "y": 72.83273243904114}, {"x": 110.97252583503723, "y": 51.15715354681015}, {"x": 19.203872233629227, "y": 37.71118628978729}, {"x": 30.421346753835678, "y": 37.16629707813263}, {"x": 39.97739166021347, "y": 40.400769114494324}, {"x": 49.04381215572357, "y": 45.61803185939789}, {"x": 56.723132252693176, "y": 53.29851043224335}, {"x": 56.084791123867035, "y": 46.76303869485855}, {"x": 65.71077406406403, "y": 46.11447751522064}, {"x": 75.0335134267807, "y": 46.96638810634613}, {"x": 83.45012521743774, "y": 51.2211879491806}, {"x": 90.43923103809357, "y": 59.06730401515961}, {"x": 57.634793639183044, "y": 60.69884181022644}, {"x": 56.081126153469086, "y": 72.10934686660767}, {"x": 54.33526223897934, "y": 84.50968235731125}, {"x": 52.046833634376526, "y": 99.18286514282227}, {"x": 43.99328297376633, "y": 99.4484453201294}, {"x": 49.026043593883514, "y": 104.39998996257782}, {"x": 55.01739031076431, "y": 107.4150230884552}, {"x": 59.81684720516205, "y": 105.90647208690643}, {"x": 64.0739357471466, "y": 101.74840021133423}, {"x": 27.087938010692596, "y": 55.09133046865463}, {"x": 34.27653384208679, "y": 55.21461433172226}, {"x": 42.54729217290878, "y": 57.39108908176422}, {"x": 48.454275488853455, "y": 61.82331371307373}, {"x": 41.1783966422081, "y": 61.66855639219284}, {"x": 34.177026987075806, "y": 60.50751310586929}, {"x": 59.51743584871292, "y": 62.6235288977623}, {"x": 65.59782898426056, "y": 61.08310657739639}, {"x": 73.38363683223724, "y": 63.624403953552246}, {"x": 78.78980994224548, "y": 67.99472141265869}, {"x": 72.09050583839417, "y": 68.4499539732933}, {"x": 64.50075209140778, "y": 66.9551106095314}, {"x": 34.73506963253021, "y": 112.76923489570618}, {"x": 42.499454855918884, "y": 113.63057744503021}, {"x": 49.24748706817627, "y": 114.29424142837524}, {"x": 55.50651115179062, "y": 115.7518025636673}, {"x": 62.20475721359253, "y": 116.19831931591034}, {"x": 66.55372452735901, "y": 117.66684019565582}, {"x": 71.82598805427551, "y": 118.23190891742706}, {"x": 64.1076112985611, "y": 127.00456368923187}, {"x": 58.63429069519043, "y": 129.7624831199646}, {"x": 52.65244817733765, "y": 129.39003217220306}, {"x": 47.463079154491425, "y": 127.39845669269562}, {"x": 42.28087282180786, "y": 123.10187613964081}, {"x": 38.320072412490845, "y": 114.88845479488373}, {"x": 49.383214592933655, "y": 119.21215355396271}, {"x": 55.4133198261261, "y": 120.68333745002747}, {"x": 60.75339984893799, "y": 121.00764191150665}, {"x": 67.49378943443298, "y": 118.56007528305054}, {"x": 59.50732082128525, "y": 124.34931361675262}, {"x": 54.06349813938141, "y": 124.39959263801575}, {"x": 48.627867102622986, "y": 122.32591986656189}]
\ No newline at end of file

test/tests/NetInput.test.ts

+import * as tf from '@tensorflow/tfjs-core';
+
+import { NetInput } from '../../src/NetInput';
+import { bufferToImage, createCanvasFromMedia } from '../../src/utils';
+import { expectAllTensorsReleased, tensor3D } from '../utils';
+
+
+describe('NetInput', () => {
+
+  let imgEl: HTMLImageElement, canvasEl: HTMLCanvasElement
+
+  beforeAll(async () => {
+    const img = await (await fetch('base/test/images/face1.png')).blob()
+    imgEl = await bufferToImage(img)
+    canvasEl = createCanvasFromMedia(imgEl)
+  })
+
+  describe('no memory leaks', () => {
+
+    describe('constructor', () => {
+
+      it('single image element', async () => {
+        await expectAllTensorsReleased(() => {
+          const net = new NetInput([imgEl])
+          net.dispose()
+        })
+      })
+
+      it('multiple image elements', async () => {
+        await expectAllTensorsReleased(() => {
+          const net = new NetInput([imgEl, imgEl, imgEl])
+          net.dispose()
+        })
+      })
+
+      it('single tf.Tensor3D', async () => {
+        const tensor = tensor3D()
+
+        await expectAllTensorsReleased(() => {
+          const net = new NetInput([tensor])
+          net.dispose()
+        })
+
+        tensor.dispose()
+      })
+
+      it('multiple tf.Tensor3Ds', async () => {
+        const tensors = [tensor3D(), tensor3D(), tensor3D()]
+
+        await expectAllTensorsReleased(() => {
+          const net = new NetInput(tensors)
+          net.dispose()
+        })
+
+        tensors.forEach(t => t.dispose())
+      })
+    })
+
+    describe('toBatchTensor', () => {
+
+      it('single image element', async () => {
+        await expectAllTensorsReleased(() => {
+          const net = new NetInput([imgEl])
+          const batchTensor = net.toBatchTensor(100, false)
+          net.dispose()
+          batchTensor.dispose()
+        })
+      })
+
+      it('multiple image elements', async () => {
+        await expectAllTensorsReleased(() => {
+          const net = new NetInput([imgEl, imgEl, imgEl])
+          const batchTensor = net.toBatchTensor(100, false)
+          net.dispose()
+          batchTensor.dispose()
+        })
+      })
+
+      it('managed, single image element', async () => {
+        await expectAllTensorsReleased(() => {
+          const net = (new NetInput([imgEl])).managed()
+          const batchTensor = net.toBatchTensor(100, false)
+          batchTensor.dispose()
+        })
+      })
+
+      it('managed, multiple image elements', async () => {
+        await expectAllTensorsReleased(() => {
+          const net = (new NetInput([imgEl, imgEl, imgEl])).managed()
+          const batchTensor = net.toBatchTensor(100, false)
+          batchTensor.dispose()
+        })
+      })
+
+    })
+
+  })
+
+})

test/tests/e2e/faceRecognitionNet.test.ts

@@ -9,7 +9,7 @@ describe('faceRecognitionNet', () => {
     const weights = new Float32Array(await res.arrayBuffer())
     faceRecognitionNet = faceapi.faceRecognitionNet(weights)
 
-    const img = await (await fetch('base/test/images/face.png')).blob()
+    const img = await (await fetch('base/test/images/face1.png')).blob()
     imgEl = await faceapi.bufferToImage(img)
     faceDescriptor = await (await fetch('base/test/data/faceDescriptor.json')).json()
   })

test/tests/toNetInput.test.ts

 import { NetInput } from '../../src/NetInput';
 import { toNetInput } from '../../src/toNetInput';
+import { bufferToImage, createCanvasFromMedia } from '../../src/utils';
+import { createFakeHTMLVideoElement } from '../utils';
 
 describe('toNetInput', () => {
 
+  let imgEl: HTMLImageElement, canvasEl: HTMLCanvasElement
+
+  beforeAll(async () => {
+    const img = await (await fetch('base/test/images/face1.png')).blob()
+    imgEl = await bufferToImage(img)
+    canvasEl = createCanvasFromMedia(imgEl)
+  })
+
   describe('valid args', () => {
 
     it('from HTMLImageElement', async () => {
-      const netInput = await toNetInput(document.createElement('img'))
+      const netInput = await toNetInput(imgEl, true)
       expect(netInput instanceof NetInput).toBe(true)
-      expect(netInput.canvases.length).toEqual(1)
+      expect(netInput.batchSize).toEqual(1)
     })
 
     it('from HTMLVideoElement', async () => {
-      const videoEl = document.createElement('video')
-      spyOnProperty(videoEl, 'readyState', 'get').and.returnValue(4)
-
-      const netInput = await toNetInput(videoEl)
+      const videoEl = await createFakeHTMLVideoElement()
+      const netInput = await toNetInput(videoEl, true)
       expect(netInput instanceof NetInput).toBe(true)
-      expect(netInput.canvases.length).toEqual(1)
+      expect(netInput.batchSize).toEqual(1)
     })
 
     it('from HTMLCanvasElement', async () => {
-      const netInput = await toNetInput(document.createElement('canvas'))
+      const netInput = await toNetInput(canvasEl, true)
       expect(netInput instanceof NetInput).toBe(true)
-      expect(netInput.canvases.length).toEqual(1)
+      expect(netInput.batchSize).toEqual(1)
     })
 
     it('from HTMLImageElement array', async () => {
       const netInput = await toNetInput([
-        document.createElement('img'),
-        document.createElement('img')
-      ])
+        imgEl,
+        imgEl
+      ], true)
       expect(netInput instanceof NetInput).toBe(true)
-      expect(netInput.canvases.length).toEqual(2)
+      expect(netInput.batchSize).toEqual(2)
     })
 
     it('from HTMLVideoElement array', async () => {
       const videoElements = [
-        document.createElement('video'),
-        document.createElement('video')
+        await createFakeHTMLVideoElement(),
+        await createFakeHTMLVideoElement()
       ]
       videoElements.forEach(videoEl =>
         spyOnProperty(videoEl, 'readyState', 'get').and.returnValue(4)
       )
 
-      const netInput = await toNetInput(videoElements)
+      const netInput = await toNetInput(videoElements, true)
       expect(netInput instanceof NetInput).toBe(true)
-      expect(netInput.canvases.length).toEqual(2)
+      expect(netInput.batchSize).toEqual(2)
     })
 
     it('from HTMLCanvasElement array', async () => {
       const netInput = await toNetInput([
-        document.createElement('canvas'),
-        document.createElement('canvas')
-      ])
+        canvasEl,
+        canvasEl
+      ], true)
       expect(netInput instanceof NetInput).toBe(true)
-      expect(netInput.canvases.length).toEqual(2)
+      expect(netInput.batchSize).toEqual(2)
     })
 
     it('from mixed media array', async () => {
-      const videoEl = document.createElement('video')
-      spyOnProperty(videoEl, 'readyState', 'get').and.returnValue(4)
-
       const netInput = await toNetInput([
-        document.createElement('img'),
-        document.createElement('canvas'),
-        videoEl
-      ])
+        imgEl,
+        canvasEl,
+        await createFakeHTMLVideoElement()
+      ], true)
       expect(netInput instanceof NetInput).toBe(true)
-      expect(netInput.canvases.length).toEqual(3)
+      expect(netInput.batchSize).toEqual(3)
     })
 
   })
@@ -81,7 +86,7 @@ describe('toNetInput', () => {
       } catch (error) {
           errorMessage = error.message;
       }
-      expect(errorMessage).toBe('toNetInput - expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement, or to be an element id')
+      expect(errorMessage).toBe('toNetInput - expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement | tf.Tensor3D, or to be an element id')
     })
 
     it('empty array', async () => {
@@ -101,7 +106,7 @@ describe('toNetInput', () => {
       } catch (error) {
           errorMessage = error.message;
       }
-      expect(errorMessage).toBe('toNetInput - at input index 1: expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement, or to be an element id')
+      expect(errorMessage).toBe('toNetInput - at input index 1: expected media to be of type HTMLImageElement | HTMLVideoElement | HTMLCanvasElement | tf.Tensor3D, or to be an element id')
     })
 
   })

test/utils.ts

+import * as tf from '@tensorflow/tfjs-core';
+
 export function zeros(length: number): Float32Array {
   return new Float32Array(length)
 }
@@ -9,3 +11,22 @@ export function ones(length: number): Float32Array {
 export function expectMaxDelta(val1: number, val2: number, maxDelta: number) {
   expect(Math.abs(val1 - val2)).toBeLessThan(maxDelta)
 }
+
+export async function createFakeHTMLVideoElement() {
+  const videoEl = document.createElement('video')
+  videoEl.muted = true
+  videoEl.src = 'base/test/media/video.mp4'
+  await videoEl.pause()
+  await videoEl.play()
+  return videoEl
+}
+
+export async function expectAllTensorsReleased(fn: () => any) {
+  const numTensorsBefore = tf.memory().numTensors
+  await fn()
+  expect(tf.memory().numTensors - numTensorsBefore).toEqual(0)
+}
+
+export function tensor3D() {
+  return tf.tensor3d([[[0]]])
+}
\ No newline at end of file