Implementing a 3D CNN for Gesture Recognition

Now that you know how the TwentyBN gesture recognition system works, you can now implement your own End-to-End 3D CNN architecture for gesture recognition.

We have created a Gesture Recognition repository on Github where you can find code that will serve as a starting point to create your own gesture recognition model. You will find the instructions on how to do this in the Github repository.

If your model performs really well on the validation set, we encourage you to run your model on the test set and to submit your results to TwentyBN’s leaderboard. Instructions on how to submit your results are given in the Jester Dataset website.