Reproduce “BBAND Index: A No-Reference Banding Artifact Predictor” 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

Zhengzhong Tu, Jessie Lin, Yilin Wang, Balu Adsumilli, Alan C. Bovik

BBAND estimates video-banding visibility without a reference or subjective ratings at prediction time. It detects banding edges, combines edge strength and perceptual masking in a pixel-level map, then pools estimates into frame and video scores. On six 720p, 30-fps VP9 clips, its Spearman correlation with viewers’ mean opinion scores was 0.9330, versus 0.8689 for Wang’s metric; results are limited to that dataset.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.