Practical Full Resolution Learned Lossless Image Compression

CVPR 2019 Fabian MentzerEirikur AgustssonMichael TschannenRadu TimofteLuc Van Gool

We propose the first practical learned lossless image compression system, L3C, and show that it outperforms the popular engineered codecs, PNG, WebP and JPEG 2000. At the core of our method is a fully parallelizable hierarchical probabilistic model for adaptive entropy coding which is optimized end-to-end for the compression task... (read more)

PDF Abstract CVPR 2019 PDF CVPR 2019 Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Image Compression ImageNet32 JPEG2000 bpsp 6.35 # 3
Image Compression ImageNet32 PNG bpsp 6.42 # 4
Image Compression ImageNet32 L3C bpsp 4.76 # 2

Methods used in the Paper