Stereoscopic video quality assessment method based on motion saliency
A technology for stereoscopic video and quality evaluation, which is applied in the field of image processing and can solve the problem of not being able to reflect the movement of three-dimensional scenes well.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Publication Date
- 2020-06-30
- Estimated Expiration
- Not applicable · inactive patent
Smart Images

Figure 1 
Figure 2 
Figure 3
Abstract
Description
technical field
[0001] The invention belongs to the field of image processing, and relates to the improvement and optimization of a stereoscopic video quality evaluation method, in particular to the application of significant motion segmentation and structural similarity in the objective evaluation of stereoscopic video quality. Specifically, it involves a method of stereoscopic video quality assessment based on motion saliency. Background technique
[0002] With the development of 3D technology in recent years, stereoscopic video is gradually becoming the main medium of visual information dissemination. However, network transmission bandwidth resources are limited, which makes the compression of stereoscopic video an inevitable choice, and practice has proved that asymmetric compression of stereoscopic video can obtain higher compression efficiency. At the same time, compression will inevitably affect the quality of stereoscopic video, and the quality of stereoscopic video...
Examples
Embodiment Construction
[0060] The invention proposes a stereoscopic video quality evaluation method based on motion salience and binocular fusion. This method first divides the image into regions according to the motion saliency, and then assigns the corresponding saliency weights to the segmented regions according to the size of the motion salience; secondly, the SSIM image quality evaluation method is used to evaluate the segmented regions to obtain local quality, and then weight the local quality of the corresponding region with the weight of motion salience to obtain the global quality of a single frame image; finally, combine the binocular competition and binocular suppression phenomena in the human visual system to assign corresponding weights to the left and right viewpoints to obtain the stereoscopic video quality. objective quality.
[0061] Specific steps are as follows:
[0062] Firstly, the corresponding motion information is estimated from the two adjacent frames (frame t and frame t+1...