human/TODO.md

93 lines
2.4 KiB
Markdown

# To-Do list for Human library
## Work in Progress
<br>
### Models
#### Hand
- Evaluate and switch default default model from `handdetect` to `handtrack`
#### Body
- Implement new variations of `BlazePose` models
- Add virtual box frame caching to `MoveNet`
#### Face
- Reimplement `BlazeFace`, `FaceMesh`, `Iris` with new pipeline and frame caching
<br>
### Backends
#### WebGL
- Optimize shader packing for WebGL backend:
<https://github.com/tensorflow/tfjs/issues/5343>
#### WASM
- Backend WASM incorrect handling of `int32` tensors
<https://github.com/tensorflow/tfjs/issues/5641>
#### WebGPU
Implementation of WebGPU backend
Experimental support only until support is officially added in Chromium
- Evaluate WGSL vs GLSL for WebGPU
- Backend WebGPU missing kernel ops
<https://github.com/tensorflow/tfjs/issues/5496>
- Backend WebGPU incompatible with web workers
<https://github.com/tensorflow/tfjs/issues/5467>
- Backend WebGPU incompatible with sync read calls
<https://github.com/tensorflow/tfjs/issues/5468>
<br>
### Exploring
- Optical Flow: <https://docs.opencv.org/3.3.1/db/d7f/tutorial_js_lucas_kanade.html>
- TFLite Models: <https://js.tensorflow.org/api_tflite/0.0.1-alpha.4/>
<br><hr><br>
## Known Issues
<br>
### Face Detection
Enhanced rotation correction for face detection is not working in NodeJS due to missing kernel op in TFJS
Feature is automatically disabled in NodeJS without user impact
- Backend NodeJS missing kernel op `RotateWithOffset`
<https://github.com/tensorflow/tfjs/issues/5473>
### Hand Detection
Enhanced rotation correction for hand detection is not working in NodeJS due to missing kernel op in TFJS
Feature is automatically disabled in NodeJS without user impact
- Backend NodeJS missing kernel op `RotateWithOffset`
<https://github.com/tensorflow/tfjs/issues/5473>
### Body Detection
MoveNet MultiPose model does not work with WASM backend due to missing F32 broadcast implementation
- Backend WASM missing F32 broadcat implementation
<https://github.com/tensorflow/tfjs/issues/5516>
### Object Detection
Object detection using CenterNet or NanoDet models is not working when using WASM backend due to missing kernel ops in TFJS
- Backend WASM missing kernel op `Mod`
<https://github.com/tensorflow/tfjs/issues/5110>
- Backend WASM missing kernel op `SparseToDense`
<https://github.com/tensorflow/tfjs/issues/4824>