Note: Please note that this repo is a fork of sense
: https://github.com/TwentyBN/sense.
Models showcased here are finetuned using the pre-trained weights and training tools available in the original repo.
Gesture control for SmartTVs
This repo demoes an RGB-based gesture control system for smart TVs. The following controls are supported:
- Play/Pause: Raise hand
- Next channel: Swipe left
- Previous channel: Swipe right
- Automatic pause when the user leaves
Try it yourself:
PYTHONPATH=./ python examples/run_smart_tv_demo.py --use_gpu
Installation steps & Troubleshooting
Please refer to the original repo for requirements & installation steps: https://github.com/TwentyBN/sense
Once everything is installed, you can check that the model is working fine by running it on the test video provided in the resources folder:
PYTHONPATH=./ python examples/run_smart_tv_demo.py --path_in=resources/smarttv_gesture_control/video_test.mp4
Model predictions should be similar to what is shown in the GIF above. In case you see a difference, check
the model framerate, displayed at the bottom-left corner; it should be close to 4 FPS. A framerate lower
than 4 indicates that your CPU isn't able to perform inference fast enough. If you have one, you may want
to run the model on the GPU instead, appending --use_gpu
to the above-mentioned command line.
License
The code is MIT but pretrained weights come with a separate license. Please check the original sense repo for more information.