Non-reference stereo image quality evaluation method based on local-to-global feature regression
A technology of stereoscopic images and global features, applied in the field of image processing, to achieve the effect of excellent performance
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0026] The network model of the present invention includes three channels (ie, left channel, right channel and fusion channel), and adopts two-step regression for training. In Step 1, the FSIM algorithm is first used to calculate the scores of the corresponding reference image and distorted image patches as labels to guide CNN to perform local regression training. After Step 1 is over, save the parameters to optimize the training of Step 2. In Step 2, the feature maps obtained from the left and right channels are concatenated with the fusion channel, and then global regression is performed based on the model of Step 1 by using DMOS as the label.
[0027] The experiment of the present invention is carried out on the publicly available LIVE 3D image database. The LIVE 3D image database includes two separate databases, phase-I and phase-II. The stereoscopic images are presented as plane images from the left and right viewpoints, and the size is 360×640. Among them, phase-I cont...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com