Feedback Recurrent Autoencoder for Video Compression
This work addresses video compression for streaming applications, offering incremental improvements over existing methods.
The authors tackled video compression by proposing a new autoencoder-based network architecture that achieves state-of-the-art MS-SSIM/rate performance on the UVG dataset, outperforming both learned and classical methods like H.265 and H.264 in streaming-relevant rate ranges.
Recent advances in deep generative modeling have enabled efficient modeling of high dimensional data distributions and opened up a new horizon for solving data compression problems. Specifically, autoencoder based learned image or video compression solutions are emerging as strong competitors to traditional approaches. In this work, We propose a new network architecture, based on common and well studied components, for learned video compression operating in low latency mode. Our method yields state of the art MS-SSIM/rate performance on the high-resolution UVG dataset, among both learned video compression approaches and classical video compression methods (H.265 and H.264) in the rate range of interest for streaming applications. Additionally, we provide an analysis of existing approaches through the lens of their underlying probabilistic graphical models. Finally, we point out issues with temporal consistency and color shift observed in empirical evaluation, and suggest directions forward to alleviate those.