A compression-distortion-oriented stereoscopic video quality objective evaluating method

Abstract

The invention discloses a compression-distortion-oriented stereoscopic video quality objective evaluating method comprising: dividing stereoscopic video quality into a left and right visual point quality section and a depth perception quality section. In the left and right visual point quality evaluating section, in view of a characteristic that a video image may generate a block effect and blur due to compression distortion, gradient information is extracted to be used as an image feature; meanwhile in view of a time-space domain visual characteristic of human eyes, left visual point quality and right visual point quality are obtained. In the depth perception quality evaluating section, by three dimensional wavelet transform, a low frequency component in a three dimensional parallax spatial diagram is extracted and the quality of the low frequency component is used as the depth perception quality of the stereoscopic video. At last, the left visual point quality and the right visual point quality are combined with the depth perception quality to obtain a final distortion stereoscopic video quality. The method has advantages of completing an object quality evaluation on the stereoscopic video from the view of the left and right visual point plane video quality and the view of the stereoscopic depth perception, and thereby effectively improving the correlation between an object evaluation result and a subject perception.

Description

technical field [0001] The invention relates to a video quality evaluation method, in particular to an objective evaluation method for compression distortion-oriented stereoscopic video quality. Background technique [0002] With the rapid development of video coding technology and display technology, various video systems have been widely used and concerned, and have gradually become the research focus in the field of information processing. Stereoscopic video has gained more and more people's favor because of its outstanding viewing experience, and the application of its related technologies has been integrated into current social life in large quantities, such as stereoscopic TV, stereoscopic film, naked eye 3D, etc. However, in the process of capturing, compressing, encoding, transmitting, and displaying stereoscopic video, various degrees and types of distortion will inevitably be introduced due to a series of uncontrollable factors. Therefore, how to accurately and ef...

AI Technical Summary

Problems solved by technology

At present, most stereoscopic video quality evaluation methods simply use the quality evaluation methods of planar video to evaluate the quality of left-viewpoint video and right-viewpoint video respectively. The impact of depth perception on the quality of stereoscopic video in stereoscopic video, so the accuracy of objective evaluation results is poor

There are also some methods that take into account the relationship between binoculars, but the weighting process between the left viewpoint and the right viewpoint is unreasonable, and it is difficult to accurately describe the perception characteristics of the human eye for stereoscopic video, and most of the current methods for evaluating the quality of stereoscopic video Time-domain weighting processing is just a simple average weighting, but in fact, the human eye's time-domain perception of stereoscopic video is not just a simple average weighting

In summary, the existing objective evaluation methods for stereoscopic video quality cannot accurately reflect the perception characteristics of human eyes, and the objective evaluation results are not accurate enough.

Images