A method and system for binocular image super-resolution reconstruction based on cross-scale disparity prior.
By employing a binocular image super-resolution reconstruction method based on cross-scale parallax priors, and utilizing cross-view interaction and cross-scale parallax attention modules, combined with a cascaded dynamic upsampling network, the problem of insufficient parallax information utilization is solved, achieving efficient and high-quality image restoration.
CN116862763BActive Publication Date: 2026-06-30YIBIN GREAT TECH CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- YIBIN GREAT TECH CO LTD
- Filing Date
- 2023-04-27
- Publication Date
- 2026-06-30
AI Technical Summary
Technical Problem
Existing stereo vision methods fail to fully utilize parallax information, resulting in poor super-resolution reconstruction of binocular images.
Method used
A binocular image super-resolution reconstruction method based on cross-scale parallax prior is adopted. Feature maps are fused through cross-view interaction and cross-scale parallax attention modules, and high-quality super-resolution images are generated by using a cascaded dynamic upsampling reconstruction network.
Benefits of technology
It generates more realistic and higher quality super-resolution images in a shorter runtime, overcoming the problem of insufficient utilization of parallax information in traditional methods and improving image restoration results.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN116862763B_ABST
Abstract
This invention discloses a method and system for binocular image super-resolution reconstruction based on cross-scale disparity prior, comprising: S1: acquiring the original left feature map and the original right feature map corresponding to the low-resolution left and right binocular images; S2: using a binocular attention module to perform cross-view interaction on the original left feature map and the original right feature map respectively, to obtain the interacted left feature map and the interacted right feature map; S3: inputting the interacted left feature map and the interacted right feature map into the cross-scale disparity attention module, and fusing them to obtain an aggregated feature map; S4: inputting the aggregated feature map into a cascaded dynamic upsampling reconstruction network to obtain the reconstructed super-resolution binocular image. This invention can fully utilize disparity information and obtain more realistic and higher-quality super-resolution images in a shorter running time.
Need to check novelty before this filing date? Find Prior Art