Learning to compress videos without computing motion

This repo holds the code for the paper:

Chen, M., Goodall, T., Patney, A., & Bovik, A. C. (2022). Learning to compress videos without computing motion. Signal Processing: Image Communication, 116633. [arXiv]

Test

TODO

Experimental Results

Evaluation results on the UVG dataset and HEVC Class B (1080p) and Class E (720p). Please refer to our paper for more experimental results.

We also provide the data of our model in file RDCurve.

Citation

If you find our paper useful, please cite:

@article{chen2022learning, title={Learning to compress videos without computing motion}, author={Chen, Meixu and Goodall, Todd and Patney, Anjul and Bovik, Alan C}, journal={Signal Processing: Image Communication}, pages={116633}, year={2022}, publisher={Elsevier} }

Contact

You can contact Meixu Chen by sending mail to [email protected]