Overfitted image coding at reduced complexity Article Swipe

PDF

Related Concepts

Coding (social sciences) Computer science Image (mathematics) Artificial intelligence Pattern recognition (psychology) Computer vision Mathematics Statistics

Théophile Blard , Théo Ladune , Pierrick Philippe , Gordon Clare , Xiaoran Jiang , Olivier Déforges ·

YOU? · · 2024 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2403.11651 · OA: W4392974110

Overfitted image codecs offer compelling compression performance and low decoder complexity, through the overfitting of a lightweight decoder for each image. Such codecs include Cool-chic, which presents image coding performance on par with VVC while requiring around 2000 multiplications per decoded pixel. This paper proposes to decrease Cool-chic encoding and decoding complexity. The encoding complexity is reduced by shortening Cool-chic training, up to the point where no overfitting is performed at all. It is also shown that a tiny neural decoder with 300 multiplications per pixel still outperforms HEVC. A near real-time CPU implementation of this decoder is made available at https://orange-opensource.github.io/Cool-Chic/.