VCT: A Video Compression Transformer
Fabian Mentzer, George Toderici, David Minnen, Sung-Jin Hwang, Sergi, Caelles, Mario Lucic, Eirikur Agustsson

TL;DR
This paper introduces VCT, a transformer-based approach for neural video compression that simplifies architecture by modeling frame dependencies directly, outperforming previous methods on standard datasets.
Contribution
The paper presents a novel transformer-based model for video compression that eliminates complex priors and achieves superior performance.
Findings
Outperforms previous video compression methods on standard datasets
Learns to handle complex motion patterns from data
Easy to implement and open-sourced
Abstract
We show how transformers can be used to vastly simplify neural video compression. Previous methods have been relying on an increasing number of architectural biases and priors, including motion prediction and warping operations, resulting in complex models. Instead, we independently map input frames to representations and use a transformer to model their dependencies, letting it predict the distribution of future representations given the past. The resulting video compression transformer outperforms previous methods on standard video compression data sets. Experiments on synthetic data show that our model learns to handle complex motion patterns such as panning, blurring and fading purely from data. Our approach is easy to implement, and we release code to facilitate future research.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
Taxonomy
TopicsAdvanced Vision and Imaging · Advanced Image Processing Techniques · Image and Signal Denoising Methods
