search

guillaume-chevalier / LSTM-Human-Activity-Recognition

Human Activity Recognition example using TensorFlow on smartphone sensors dataset and an LSTM RNN. Classifying the type of movement amongst six activity categories - Guillaume Chevalier
MIT License
3.33k stars 935 forks source link

Realtime classification #10

Closed eighteight closed 7 years ago

eighteight commented 7 years ago

Hi!

Thanks for posting this code. Do you think this approach will work for realtime classification of the human activity?

guillaume-chevalier commented 7 years ago

The code can be exported to C++ with the TensorFlow library. Training may be long, but doing predictions is fast. This can run real-time if you already have at least 2.56 seconds of recorded signal available. For example, from the moment you start your app, you could have to wait 2.56 seconds only one time, then predictions could be real-time.

If this is too long, you could shorten the time window, but accuracy may (or may not?) decrease: to be tested.