Binary Probability Model for Learning Based Image Compression
This work addresses the need for more efficient image compression for applications like storage and transmission, though it appears incremental as it builds on existing learned compression systems.
The paper tackles the problem of improving learned image compression by proposing a richer probability model for latent variables, achieving 18% rate savings compared to Gaussian or Laplace models in experiments under CLIC test conditions.
In this paper, we propose to enhance learned image compression systems with a richer probability model for the latent variables. Previous works model the latents with a Gaussian or a Laplace distribution. Inspired by binary arithmetic coding , we propose to signal the latents with three binary values and one integer, with different probability models. A relaxation method is designed to perform gradient-based training. The richer probability model results in a better entropy coding leading to lower rate. Experiments under the Challenge on Learned Image Compression (CLIC) test conditions demonstrate that this method achieves 18% rate saving compared to Gaussian or Laplace models.