Full-reference video quality evaluation method based on motion estimation

Abstract

The invention belongs to the field of full-reference video quality evaluation, and particularly provides a full-reference video quality evaluation method based on motion estimation, which comprises the following steps: firstly, calculating the change of video content gradient information caused by distortion in space, and obtaining a spatial distortion score value of a video through a two-step time pooling mode; fusing the difference between the local mean value of the motion vector diagram and the standard deviation variable coefficient, the difference between the variable coefficients of the motion vectors at the same position of the adjacent frames and the spatial distortion characteristics to obtain a space-time distortion score value of the video; and finally, fusing the spatial distortion score value and the space-time distortion score value to obtain a final video quality prediction value. According to the method, the motion information is introduced into the video quality evaluation method, so that the accuracy of video quality prediction can be remarkably enhanced, and an objective video quality evaluation method with higher consistency with subjective feeling of human eyes is further obtained.

Description

technical field [0001] The invention belongs to the field of full-reference video quality assessment, and specifically provides a full-reference video quality assessment method using motion estimation. Background technique [0002] With the rapid development of network technology, more and more videos are disseminated on the Internet; during the process of shooting, encoding and transmitting videos, various distortions are often introduced, resulting in the degradation of video quality and seriously affecting the viewing of users Experience; how to accurately measure the quality of video has important practical significance for the fields of encoding, video transmission and terminal video quality enhancement. [0003] Since subjective video quality assessment requires a large number of experimenters to score videos one by one, it is time-consuming and laborious, and in actual scenarios, it is difficult to implement subjective quality assessment for each video; at the same ti...

AI Technical Summary

Problems solved by technology

[0002] With the rapid development of network technology, more and more videos are disseminated on the Internet; during the process of shooting, encoding and transmitting videos, various distortions are often introduced, resulting in the degradation of video quality and seriously affecting the viewing of users Experience; how to accurately measure the quality of video has important practical significance for the fields of encoding, video transmission and terminal video quality enhancement

[0003] Since subjective video quality assessment requires a large number of experimenters to score videos one by one, it is time-consuming and laborious, and in actual scenarios, it is difficult to implement subjective quality assessment for each video; at the same time, the objective method for measuring video quality that is often used at present, Such as PSNR, SSIM, they are quite different from the subjective experience of human eyes, and cannot evaluate the video quality well

Images