Adapting objective video quality metrics to ambient lighting

The limitations of objective metrics such as PSNR in evaluating video quality are well known to experts but less known to many users. A video tool which exploits perceptual phenomena may give improved subjective quality but lower objective performance when evaluated by such metrics. This presents a problem when describing the performance of perceptually motivated algorithms to general users. We propose and evaluate a method of extending existing objective metrics to account for perceptual factors such as viewing distance and ambient contrast. After describing the proposed algorithm we examine the variation with ambient lighting of the proposed modification. Subjective viewing tests are used to confirm the behavior of the extended objective metrics.