Stereo Matching Pyramid Search Range Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Pyramid-based stereo matching methods often fail to accurately recover thin objects due to low resolution at coarse levels, causing them to blend into the background and be lost in disparity maps.
Innovation Solution
Adaptive adjustment of the search range in each pyramid level based on the nature of the image, increasing the search range when thin objects are present to enhance the visibility of these objects in the depth map, using a correlation score to determine the disparity estimation value and improve the accuracy of the depth map.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If pyramid based methods are used for coarse to fine computations, then computational efficiency is improved, but thin objects are lost due to low resolution at coarse levels
Solution Approach 1:
The patent applies dynamics by making the search range adaptive rather than fixed. The search range dynamically adjusts based on the pyramid level being processed, expanding at coarser levels to compensate for reduced resolution. This allows the system to maintain thin object detection capability across all pyramid levels while preserving computational efficiency of the coarse-to-fine approach
Solution Approach 2:
The patent changes the search range parameter adaptively at different pyramid levels. At coarser levels where resolution is reduced and thin objects may be lost, the search range is increased to ensure these objects are still captured. This parameter adjustment resolves the contradiction by allowing the system to maintain both computational efficiency and thin object recovery accuracy
2Measurement precision
If search range is increased to capture thin objects, then thin object visibility is improved, but computational complexity increases
Solution Approach 1:
The patent applies local quality by making the search range adaptive to local image characteristics and pyramid level rather than uniformly applying a large search range everywhere. The search range is selectively increased only where and when needed (at coarser pyramid levels) to capture thin objects, rather than increasing it globally across all levels and all regions, thus managing computational complexity while improving thin object visibility
Data Source
AI summary
Techniques to determine a search range for a stereo based matching pyramid. A first disparity estimation value for a first level in a stereo based matching pyramid based on an image may be received. A search range for a second level may be determined using the first disparity estimation value. The search range based on a pyramid level of a second level may be increased. The search range may be increased based on a pyramid level of the second level. A second disparity estimation value may be selected from the search area for the second level. A depth map for the second level may be determined based on the second disparity estimation value. Other embodiments are described and claimed.