Awesome
Compressed Video Action Recognition <br>
Chao-Yuan Wu,
Manzil Zaheer,
Hexiang Hu,
R. Manmatha,
Alexander J. Smola,
Philipp Krähenbühl.
In CVPR, 2018.
[Project Page]
Overview
This is a reimplementation of CoViAR in PyTorch (the original paper uses MXNet). This code currently supports UCF-101 and HMDB-51; Charades coming soon. (This is a work in progress. Any suggestions are appreciated.)
Results
This code produces comparable or better results than the original paper:
HMDB-51: 52% (I-frame), 40% (motion vector), 43% (residuals), 59.2% (CoViAR).
UCF-101: 87% (I-frame), 70% (motion vector), 80% (residuals), 90.5% (CoViAR).
(average of 3 splits; <b>without</b> optical flow. )
Data loader
<b>We provide a python data loader that directly takes a compressed video and returns the compressed representation (I-frames, motion vectors, and residuals) as a numpy array </b>. We can thus train the model without extracting and storing all representations as image files.
In our experiments, it's fast enough so that it doesn't delay GPU training.
Please see GETTING_STARTED.md
for details and instructions.
Using CoViAR
Please see GETTING_STARTED.md
for instructions for training and inference.
Citation
If you find this model useful for your resesarch, please use the following BibTeX entry.
<pre> @inproceedings{wu2018coviar, title={Compressed Video Action Recognition}, author={Wu, Chao-Yuan and Zaheer, Manzil and Hu, Hexiang and Manmatha, R and Smola, Alexander J and Kr{\"a}henb{\"u}hl, Philipp}, booktitle={CVPR}, year={2018} } </pre>Acknowledgment
This implementation largely borrows from tsn-pytorch by yjxiong.
Part of the dataloader implementation is modified from this tutorial and FFmpeg extract_mv
example.