Google Research

Benchmarking JPEG XL lossy/lossless image compression

Optics, Photonics and Digital Technologies for Imaging Applications VI, SPIE (2020)

Abstract

JPEG XL is a practical, royalty-free codec for scalable web distribution and efficient compression of high-quality photographs. It also includes previews, progression, animation, transparency, wide gamut, and high bit depth. Experiments performed during standardization have shown the feasibility of economical storage without perceptive quality loss, lossless recompression of existing JPEG, and fast software encoders and decoders. We disclose the results of subjective and objective evaluations. Users expect faithful reproductions of ever-larger images. JPEG XL is faster to share and more economical to store: 60% savings vs. JPEG at equivalent visual quality. We quantify this impact using a subjective evaluation versus existing codecs including HEVC-HM-YUV444 and JPEG. New image codecs have to co-exist with the previous generation for several years. JPEG XL is unique in providing value for both existing JPEGs as well as new users. It includes coding tools to reduce the transmission and storage costs of JPEG by 22% while allowing byte-for-byte exact reconstruction of the original JPEG. Avoiding transcoding and additional artifacts helps to preserve our digital heritage. Applications require fast and low-power decoding. JPEG XL was designed to benefit from multicore and SIMD, and actually decodes faster than JPEG. We report the resulting speeds on ARM and x86 CPUs. To enable reproduction of these results, we open sourced the JPEG XL software in 2019.

Learn more about how we do research

We maintain a portfolio of research projects, providing individuals and teams the freedom to emphasize specific types of work