MCPcopy
hub / github.com/tensorflow/tfjs-models / PoseDetector

Interface PoseDetector

pose-detection/src/pose_detector.ts:26–51  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

24 * User-facing interface for all pose detectors.
25 */
26export interface PoseDetector {
27 /**
28 * Estimate poses for an image or video frame.
29 * @param image An image or video frame.
30 * @param config Optional. See `EstimationConfig` for available options.
31 * @param timestamp Optional. In milliseconds. This is useful when image is
32 * a tensor, which doesn't have timestamp info. Or to override timestamp
33 * in a video.
34 * @returns An array of poses, each pose contains an array of `Keypoint`s.
35 */
36 estimatePoses(
37 image: PoseDetectorInput,
38 config?: PoseNetEstimationConfig|BlazePoseTfjsEstimationConfig|
39 BlazePoseMediaPipeEstimationConfig|MoveNetEstimationConfig,
40 timestamp?: number): Promise<Pose[]>;
41
42 /**
43 * Dispose the underlying models from memory.
44 */
45 dispose(): void;
46
47 /**
48 * Reset global states in the model.
49 */
50 reset(): void;
51}

Callers 19

poseDetectionFrameFunction · 0.65
renderResultFunction · 0.65
renderResultFunction · 0.65
runFunction · 0.65
movenet_test.tsFile · 0.65
callbackFunction · 0.65
posenet_test.tsFile · 0.65
disposeMethod · 0.65
disposeFunction · 0.65
estimateMultiplePosesMethod · 0.65
estimateSinglePoseMethod · 0.65

Implementers 4

MoveNetDetectorpose-detection/src/movenet/detector.ts
PosenetDetectorpose-detection/src/posenet/detector.ts
BlazePoseTfjsDetectorpose-detection/src/blazepose_tfjs/dete
BlazePoseMediaPipeDetectorpose-detection/src/blazepose_mediapipe

Calls

no outgoing calls

Tested by

no test coverage detected