Stereo Correspondence Using Depth Map Discontinuities
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional stereo correspondence techniques face ambiguities in processing images with lack of textures or features, repeated patterns, shadows, and view-point changes, leading to difficulties in determining corresponding points in stereo images.
Innovation Solution
The use of a depth map generated by a depth sensor to assist in stereo correspondence calculations, aligning depth discontinuities with image edges and defining a search range for improved accuracy and efficiency, while suppressing false positives by weighting image edges based on correspondence to depth discontinuities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional stereo correspondence techniques are used, then processing can be performed on standard images, but accuracy deteriorates in regions without textures or features, with repeated patterns, shadows, or view-point changes
Solution Approach 1:
The patent introduces a depth map as an intermediary element that mediates between the stereo images and the correspondence calculation. The depth map provides additional depth information that helps disambiguate correspondences in challenging regions such as textureless areas, repeated patterns, and shadow regions, thereby improving both accuracy and reliability without requiring changes to the fundamental stereo matching approach
Solution Approach 2:
The patent transitions from two-dimensional image plane analysis to three-dimensional space by incorporating depth maps. This dimensional extension allows the system to use depth information to resolve ambiguities that cannot be solved using only image intensity and gradient information, particularly in regions where conventional 2D cues fail
2Measurement precision
If conventional stereo correspondence techniques are used, then processing can be completed without additional sensors, but measurement precision deteriorates due to ambiguities in textureless or patterned regions
Solution Approach 1:
The patent makes the depth map serve multiple functions: it acts as a guide for correspondence search, provides depth constraints for disambiguation, and helps define search ranges. This multi-functionality allows a single additional sensor to address multiple sources of ambiguity simultaneously, improving depth estimation accuracy while keeping the system configuration relatively simple
3Measurement precision
If exhaustive search is performed for stereo correspondence, then measurement precision may be improved, but processing time increases significantly
Solution Approach 1:
The patent performs preliminary actions by using the depth map to pre-establish correspondence relationships and define search ranges before the actual stereo matching process. This preliminary guidance from depth information allows the algorithm to focus computations only on relevant regions, significantly reducing processing time while maintaining high accuracy in identifying corresponding points
Data Source
AI summary
Depth map stereo correspondence techniques are described. In one or more implementations, a depth map generated through use of a depth sensor is leveraged as part of processing of stereo images to assist in identifying which parts of stereo images correspond to each other. For example, the depth map may be utilized to describe depth of an image scene which may be used as part of a stereo correspondence calculation. The depth map may also be utilized as part of a determination of a search range to be employed as part of the stereo correspondence calculation.


