Contextformer: A Transformer with Spatio-Channel Attention for Context Modeling in Learned Image Compression
This work improves image compression efficiency for applications like storage and transmission, though it appears incremental as it builds on existing transformer and compression frameworks.
The paper tackles the problem of improving entropy modeling in learned image compression by addressing underexploited spatio-channel dependencies and suboptimal context adaptivity, resulting in up to 11% rate savings compared to VVC VTM 16.2 and outperforming other learning-based models on PSNR and MS-SSIM metrics.
Entropy modeling is a key component for high-performance image compression algorithms. Recent developments in autoregressive context modeling helped learning-based methods to surpass their classical counterparts. However, the performance of those models can be further improved due to the underexploited spatio-channel dependencies in latent space, and the suboptimal implementation of context adaptivity. Inspired by the adaptive characteristics of the transformers, we propose a transformer-based context model, named Contextformer, which generalizes the de facto standard attention mechanism to spatio-channel attention. We replace the context model of a modern compression framework with the Contextformer and test it on the widely used Kodak, CLIC2020, and Tecnick image datasets. Our experimental results show that the proposed model provides up to 11% rate savings compared to the standard Versatile Video Coding (VVC) Test Model (VTM) 16.2, and outperforms various learning-based models in terms of PSNR and MS-SSIM.