7th IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI 2006), Colorado, United States Of America, 26 - 28 March 2006, pp.218-219
We suggest a compound full-reference stereo-video quality metric composed of two components: a monoscopic quality component and stereoscopic quality, component. While the former assesses the trivial monoscopic perceived distortions caused by blur, noise, contrast change etc., the latter assesses the perceived degradation of binocular depth cues only. We use the structural similarity index as a measure for perceptual similarity and design a multiscale algorithm for obtaining a perceptual disparity map and a stereo-similarity map to be used in the suggested metric. We verify the performance of the metric with subjective tests on distorted stereo images and coded stereo-video sequences with a final aim to build a percepually-aware feedback for a H.264 based stereo-video encoder.