fixed mtcnn face landmark positions

src/classes/FaceDetectionWithLandmarks.ts

 import { FaceDetection } from './FaceDetection';
 import { FaceLandmarks } from './FaceLandmarks';
+import { FaceLandmarks68 } from './FaceLandmarks68';
 
-export class FaceDetectionWithLandmarks {
+export class FaceDetectionWithLandmarks<TFaceLandmarks extends FaceLandmarks = FaceLandmarks68> {
   private _detection: FaceDetection
-  private _relativeLandmarks: FaceLandmarks
+  private _unshiftedLandmarks: TFaceLandmarks
 
   constructor(
     detection: FaceDetection,
-    relativeLandmarks: FaceLandmarks
+    unshiftedLandmarks: TFaceLandmarks
   ) {
     this._detection = detection
-    this._relativeLandmarks = relativeLandmarks
+    this._unshiftedLandmarks = unshiftedLandmarks
   }
 
   public get detection(): FaceDetection { return this._detection }
-  public get relativeLandmarks(): FaceLandmarks { return this._relativeLandmarks }
+  public get unshiftedLandmarks(): TFaceLandmarks { return this._unshiftedLandmarks }
 
   public get alignedRect(): FaceDetection {
     const rect = this.landmarks.align()
@@ -22,14 +23,18 @@ export class FaceDetectionWithLandmarks {
     return new FaceDetection(this._detection.score, rect.rescale(imageDims.reverse()), imageDims)
   }
 
-  public get landmarks(): FaceLandmarks {
+  public get landmarks(): TFaceLandmarks {
     const { x, y } = this.detection.box
-    return this._relativeLandmarks.shift(x, y)
+    return this._unshiftedLandmarks.shiftBy(x, y)
   }
 
-  public forSize(width: number, height: number): FaceDetectionWithLandmarks {
+  // aliases for backward compatibily
+  get faceDetection(): FaceDetection { return this.detection }
+  get faceLandmarks(): TFaceLandmarks { return this.landmarks }
+
+  public forSize(width: number, height: number): FaceDetectionWithLandmarks<TFaceLandmarks> {
     const resizedDetection = this._detection.forSize(width, height)
-    const resizedLandmarks = this._relativeLandmarks.forSize(resizedDetection.box.width, resizedDetection.box.height)
-    return new FaceDetectionWithLandmarks(resizedDetection, resizedLandmarks)
+    const resizedLandmarks = this._unshiftedLandmarks.forSize<TFaceLandmarks>(resizedDetection.box.width, resizedDetection.box.height)
+    return new FaceDetectionWithLandmarks<TFaceLandmarks>(resizedDetection, resizedLandmarks)
   }
 }
\ No newline at end of file

src/classes/FaceLandmarks.ts

@@ -25,10 +25,7 @@ export class FaceLandmarks {
     )
   }
 
-  public getShift(): Point {
-    return new Point(this._shift.x, this._shift.y)
-  }
-
+  public get shift(): Point { return new Point(this._shift.x, this._shift.y) }
   public get imageWidth(): number { return this._imgDims.width }
   public get imageHeight(): number { return this._imgDims.height }
   public get positions(): Point[] { return this._positions }
@@ -45,7 +42,7 @@ export class FaceLandmarks {
     )
   }
 
-  public shift<T extends FaceLandmarks>(x: number, y: number): T {
+  public shiftBy<T extends FaceLandmarks>(x: number, y: number): T {
     return new (this.constructor as any)(
       this.relativePositions,
       this._imgDims,
@@ -54,7 +51,7 @@ export class FaceLandmarks {
   }
 
   public shiftByPoint<T extends FaceLandmarks>(pt: Point): T {
-    return this.shift(pt.x, pt.y)
+    return this.shiftBy(pt.x, pt.y)
   }
 
   /**
@@ -76,7 +73,7 @@ export class FaceLandmarks {
         ? detection.box.floor()
         : detection
 
-      return this.shift(box.x, box.y).align()
+      return this.shiftBy(box.x, box.y).align()
     }
 
     const centers = this.getRefPointsForAlignment()

src/classes/FullFaceDescription.ts

 import { FaceDetection } from './FaceDetection';
 import { FaceDetectionWithLandmarks } from './FaceDetectionWithLandmarks';
 import { FaceLandmarks } from './FaceLandmarks';
+import { FaceLandmarks68 } from './FaceLandmarks68';
 
-export class FullFaceDescription extends FaceDetectionWithLandmarks {
+export class FullFaceDescription<TFaceLandmarks extends FaceLandmarks = FaceLandmarks68> extends FaceDetectionWithLandmarks<TFaceLandmarks> {
   private _descriptor: Float32Array
 
   constructor(
     detection: FaceDetection,
-    landmarks: FaceLandmarks,
+    unshiftedLandmarks: TFaceLandmarks,
     descriptor: Float32Array
   ) {
-    super(detection, landmarks)
+    super(detection, unshiftedLandmarks)
     this._descriptor = descriptor
   }
 
@@ -18,8 +19,8 @@ export class FullFaceDescription extends FaceDetectionWithLandmarks {
     return this._descriptor
   }
 
-  public forSize(width: number, height: number): FullFaceDescription {
+  public forSize(width: number, height: number): FullFaceDescription<TFaceLandmarks> {
     const { detection, landmarks } = super.forSize(width, height)
-    return new FullFaceDescription(detection, landmarks, this.descriptor)
+    return new FullFaceDescription<TFaceLandmarks>(detection, landmarks, this.descriptor)
   }
 }
\ No newline at end of file

src/globalApi/nets.ts

@@ -2,13 +2,13 @@ import { TNetInput } from 'tfjs-image-recognition-base';
 import { ITinyYolov2Options } from 'tfjs-tiny-yolov2';
 
 import { FaceDetection } from '../classes/FaceDetection';
+import { FaceDetectionWithLandmarks } from '../classes/FaceDetectionWithLandmarks';
 import { FaceLandmarks68 } from '../classes/FaceLandmarks68';
 import { FaceLandmark68Net } from '../faceLandmarkNet/FaceLandmark68Net';
 import { FaceLandmark68TinyNet } from '../faceLandmarkNet/FaceLandmark68TinyNet';
 import { FaceRecognitionNet } from '../faceRecognitionNet/FaceRecognitionNet';
 import { Mtcnn } from '../mtcnn/Mtcnn';
 import { MtcnnOptions } from '../mtcnn/MtcnnOptions';
-import { MtcnnResult } from '../mtcnn/MtcnnResult';
 import { SsdMobilenetv1 } from '../ssdMobilenetv1/SsdMobilenetv1';
 import { SsdMobilenetv1Options } from '../ssdMobilenetv1/SsdMobilenetv1Options';
 import { TinyFaceDetector } from '../tinyFaceDetector/TinyFaceDetector';
@@ -63,7 +63,7 @@ export const tinyYolov2 = (input: TNetInput, options: ITinyYolov2Options): Promi
  * @param options (optional, default: see MtcnnOptions constructor for default parameters).
  * @returns Bounding box of each face with score and 5 point face landmarks.
  */
-export const mtcnn = (input: TNetInput, options: MtcnnOptions): Promise<MtcnnResult[]> =>
+export const mtcnn = (input: TNetInput, options: MtcnnOptions): Promise<FaceDetectionWithLandmarks[]> =>
   nets.mtcnn.forward(input, options)
 
 /**

src/mtcnn/Mtcnn.ts

@@ -9,12 +9,12 @@ import { extractParams } from './extractParams';
 import { getSizesForScale } from './getSizesForScale';
 import { loadQuantizedParams } from './loadQuantizedParams';
 import { IMtcnnOptions, MtcnnOptions } from './MtcnnOptions';
-import { MtcnnResult } from './MtcnnResult';
 import { pyramidDown } from './pyramidDown';
 import { stage1 } from './stage1';
 import { stage2 } from './stage2';
 import { stage3 } from './stage3';
 import { NetParams } from './types';
+import { FaceDetectionWithLandmarks } from '../classes/FaceDetectionWithLandmarks';
 
 export class Mtcnn extends NeuralNetwork<NetParams> {
 
@@ -25,7 +25,7 @@ export class Mtcnn extends NeuralNetwork<NetParams> {
   public async forwardInput(
     input: NetInput,
     forwardParams: IMtcnnOptions = {}
-  ): Promise<{ results: MtcnnResult[], stats: any }> {
+  ): Promise<{ results: FaceDetectionWithLandmarks[], stats: any }> {
 
     const { params } = this
 
@@ -101,7 +101,7 @@ export class Mtcnn extends NeuralNetwork<NetParams> {
     const out3 = await stage3(inputCanvas, out2.boxes, scoreThresholds[2], params.onet, stats)
     stats.total_stage3 = Date.now() - ts
 
-    const results = out3.boxes.map((box, idx) => new MtcnnResult(
+    const results = out3.boxes.map((box, idx) => new FaceDetectionWithLandmarks(
       new FaceDetection(
         out3.scores[idx],
         new Rect(
@@ -116,8 +116,8 @@ export class Mtcnn extends NeuralNetwork<NetParams> {
         }
       ),
       new FaceLandmarks5(
-        out3.points[idx].map(pt => pt.div(new Point(width, height))),
-        { width, height }
+        out3.points[idx].map(pt => pt.sub(new Point(box.left, box.top)).div(new Point(box.width, box.height))),
+        { width: box.width, height: box.height }
       )
     ))
 
@@ -127,7 +127,7 @@ export class Mtcnn extends NeuralNetwork<NetParams> {
   public async forward(
     input: TNetInput,
     forwardParams: IMtcnnOptions = {}
-  ): Promise<MtcnnResult[]> {
+  ): Promise<FaceDetectionWithLandmarks[]> {
     return (
       await this.forwardInput(
         await toNetInput(input),
@@ -139,7 +139,7 @@ export class Mtcnn extends NeuralNetwork<NetParams> {
   public async forwardWithStats(
     input: TNetInput,
     forwardParams: IMtcnnOptions = {}
-  ): Promise<{ results: MtcnnResult[], stats: any }> {
+  ): Promise<{ results: FaceDetectionWithLandmarks[], stats: any }> {
     return this.forwardInput(
       await toNetInput(input),
       forwardParams

src/mtcnn/MtcnnResult.ts

-import { FaceDetection } from '../classes/FaceDetection';
-import { FaceDetectionWithLandmarks } from '../classes/FaceDetectionWithLandmarks';
-import { FaceLandmarks5 } from '../classes/FaceLandmarks5';
-
-export class MtcnnResult extends FaceDetectionWithLandmarks {
-
-  // aliases for backward compatibily
-  get faceDetection(): FaceDetection { return this.detection }
-  get faceLandmarks(): FaceLandmarks5 { return this.faceLandmarks }
-}
\ No newline at end of file

src/ssdMobilenetv1/SsdMobilenetv1.ts

@@ -46,7 +46,7 @@ export class SsdMobilenetv1 extends NeuralNetwork<NetParams> {
 
   public async locateFaces(
     input: TNetInput,
-    options: ISsdMobilenetv1Options
+    options: ISsdMobilenetv1Options = {}
   ): Promise<FaceDetection[]> {
 
     const { maxResults, minConfidence } = new SsdMobilenetv1Options(options)