# An Improved Upper Bound on the Rate-Distortion Function of Images

arXiv: https://arxiv.org/abs/2309.02574 · HTML: https://mlsanity.com/paper/2309.02574/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Sep 2023.

## Built on by (most cited first, 1 of 1)

- [Efficient training for compact compression models via sequential distillation](https://mlsanity.com/paper/2601.05639/) (2026)

## Abstract (arXiv)

Recent work has shown that Variational Autoencoders (VAEs) can be used to upper-bound the information rate-distortion (R-D) function of images, i.e., the fundamental limit of lossy image compression. In this paper, we report an improved upper bound on the R-D function of images implemented by (1) introducing a new VAE model architecture, (2) applying variable-rate compression techniques, and (3) proposing a novel ourfunction to stabilize training. We demonstrate that at least 30% BD-rate reduction w.r.t. the intra prediction mode in VVC codec is achievable, suggesting that there is still great potential for improving lossy image compression. Code is made publicly available at https://github.com/duanzhiihao/lossy-vae.