Users prefer Guetzli JPEG over same-sized libjpeg
arXiv:1703.04416v14 citations
Originality Synthesis-oriented
AI Analysis
This addresses image compression quality for users, but is incremental as it builds on existing metrics.
The study tackled the problem of JPEG image quality perception by comparing human preferences between libjpeg and the new Guetzli encoder at matched file sizes, finding that 75% of ratings favored Guetzli.
We report on pairwise comparisons by human raters of JPEG images from libjpeg and our new Guetzli encoder. Although both files are size-matched, 75% of ratings are in favor of Guetzli. This implies the Butteraugli psychovisual image similarity metric which guides Guetzli is reasonably close to human perception at high quality levels. We provide access to the raw ratings and source images for further analysis and study.