Reproduce “FMT: Lossless Image Compression in a Fixed Number of Parallel Steps” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ammar Ali
The authors present FMT, a family of lossless image codecs that decode images through parallel steps. Pixel-tree modes recover a coarse grid, then predict independent groups of remaining pixels and encode corrections with state-carrying entropy-coder lanes. This bounds sequential steps as image size grows. On Kodak, FMT 512×4 is reported 10.2% smaller and 8.8× faster than FNLIC in a comparison using one A100.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.