CAMBI: Contrast-aware Multiscale Banding Index

TL;DR

This paper introduces CAMBI, a no-reference index for detecting video banding artifacts, based on human visual system insights, which correlates well with subjective perception and aids in quality assessment.

Contribution

The paper develops CAMBI, a novel contrast-aware multiscale banding index that predicts banding visibility using minimal hyperparameters and human visual perception principles.

Findings

CAMBI correlates strongly with subjective banding perception.

CAMBI outperforms existing no-reference banding metrics.

The study highlights the importance of contrast sensitivity in banding detection.

Abstract

Banding artifacts are artificially-introduced contours arising from the quantization of a smooth region in a video. Despite the advent of recent higher quality video systems with more efficient codecs, these artifacts remain conspicuous, especially on larger displays. In this work, a comprehensive subjective study is performed to understand the dependence of the banding visibility on encoding parameters and dithering. We subsequently develop a simple and intuitive no-reference banding index called CAMBI (Contrast-aware Multiscale Banding Index) which uses insights from Contrast Sensitivity Function in the Human Visual System to predict banding visibility. CAMBI correlates well with subjective perception of banding while using only a few visually-motivated hyperparameters.