Video redirection quality evaluation method based on space-time saliency classification and fusion

Abstract

The invention discloses a video redirection quality evaluation method based on space-time saliency classification and fusion. The method comprises the following steps: constructing a video classification model according to space-time saliency of videos, and dividing the videos into four classes according to motion information and foreground information of original videos; extracting salient information, edge features, foreground information, motion features and the like of the original video, and evaluating the quality of the redirected video by adopting four spatio-temporal indexes, namely perceptual geometric distortion, edge group similarity, time continuity similarity distortion and important target time distortion; adopting different self-adaptive weight endowing methods for different types of videos, and fusing the quality scores of the four spatio-temporal indexes to obtain the overall objective quality of the redirected video. An index weight adaptive fusion mode is adopted for the classified videos, different space-time characteristics are extracted for different types of videos, different index weight adaptive fusion methods are adopted, the characteristics of the videos are fully considered, and the performance of an objective evaluation algorithm is greatly ensured.

Description

technical field [0001] The invention belongs to the technical field of image evaluation, and more particularly relates to a video redirection quality evaluation method based on space-time saliency classification and fusion. Background technique [0002] With the rapid popularization of multimedia display devices such as LCD TVs, tablets, laptops and smart phones, and the explosive growth of video data, people can watch media videos anytime and anywhere. However, due to different application requirements, terminal display devices often do not have a uniform resolution and aspect ratio, and the same original video will be stretched or squeezed to different degrees when displayed on terminals with different resolutions and aspect ratios. Not only will it cause a waste of display screen space, but it will also seriously affect the user's viewing experience. In order to avoid this phenomenon, it is necessary to adjust the video content, adaptively adjust its resolution or aspect...

AI Technical Summary

Problems solved by technology

Due to the disadvantages of subjective evaluation: 1) The testers participating in the subjective evaluation usually directly judge the quality of the redirected video based on their own subjective feelings, and it is difficult to quantitatively analyze and evaluate the impact of various distortions on the quality of the reconstructed video during the redirection process. The evaluation process and results are often difficult to directly apply to the performance improvement of the redirection algorithm; 2) Subjective evaluation requires a large number of testers to vote repeatedly on the combination of redirected videos. Influenced by factors such as observation environment and observation environment, it is difficult to directly embed the results of subjective evaluation into emerging real-time video applications, and the portability is low.

Although this method can evaluate the overall quality of the redirected video, it is difficult to evaluate the geometric distortion of the video space well, and cannot evaluate the temporal distortion of the target still video.

To sum up, the existing objective evaluation algorithms for video redirection quality have single evaluation indicators, do not fully consider the characteristics of HVS, are difficult to apply to different types of videos, and often use fixed index weight fusion methods, resulting in evaluation scores Difficult to conform to the results of human subjective perception

Images