20BN-RealtimeNet is an efficient implementation of a state-of-the-art Deep Learning neural network for video understanding. The inference code is open source and available on GitHub under MIT license: https://github.com/TwentyBN/20bn-realtimenet
We provide pre-trained model files for the above inference code for download on this page. The model has been pre-trained on several million video clips that we collected using our unqiue CrowdActing (tm) approach.
The model files are not open source.
We have three different license types available.