microsoft / human-pose-estimation.pytorch

The project is an official implement of our ECCV2018 paper "Simple Baselines for Human Pose Estimation and Tracking(https://arxiv.org/abs/1804.06208)"
MIT License
2.94k stars 605 forks source link

Is there any mobile implementation? Androis / iOS? #164

Open yonatanbitton opened 3 years ago

yonatanbitton commented 3 years ago

Is there any mobile implementation? Androis / iOS? Thanks

tucachmo2202 commented 3 years ago

Hi @yonatanbitton, I think you may convert the model to tflite or coreml format for deploying in Android and IOS (example: torch->onnx->tensorflow->tensorflow lite), then use the preprocessing, postprocessing code to inference. However, I don't highly recommend you, the two-stage model require a detector and it is not quick enough for realtime in mobile. Anyway, I haven't try it yet, so you can try for experience then let me know about the time. Thanks!