We provide PyTorch implementations for Temporal Segmentation Network with Dynamic Image for Action Recognition .
The code was written by Daniel Kurniadi, and supported by Wang Zheng and Liu Siying.
This PyTorch implementation produces results comparable to or better than our original Torch software. If you would like to reproduce the same results as in the papers, check out the original Temporal Segmentation Network and Dynamic Imagenet code base.
Note: The current software works well with PyTorch 1.0.1