Two-Stage Disparity Computation for Image Precision and Speed
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computer vision technologies face challenges in achieving both precision and speed in disparity computation, particularly when dealing with large images, as the disparity search range and image size affect computation speed and precision, leading to a trade-off between the two.
Innovation Solution
A two-stage disparity computation method is employed, where images are downscaled initially to reduce the disparity search range, followed by upsampling and subsequent disparity computation on the initial disparity map to enhance precision and speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If disparity computation is performed on full-sized images, then measurement precision is improved, but productivity deteriorates due to increased computational load
Solution Approach 1:
The disparity computation process is segmented into two distinct stages: a first stage performing computation on downsampled images to obtain initial disparity information, and a second stage performing computation on full-sized images using the initial disparity as a reference. This segmentation allows each stage to operate optimally - the first stage provides coarse disparity quickly, while the second stage refines precision without full computational burden
Solution Approach 2:
The first-stage disparity computation on downsampled images serves as a preliminary action that prepares initial disparity information before the second-stage computation. This preliminary disparity map is then upsampled and used as a reference for the second stage, eliminating the need for exhaustive search in the second stage and significantly improving computation speed while maintaining precision
2Productivity
If image size is reduced for computation, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The downsampled disparity computation serves as a preliminary action that obtains initial disparity information quickly. This initial disparity is then upsampled to the target resolution and used as a reference for the second-stage computation on full-sized images, ensuring that the final precision matches full-resolution computation while benefiting from the speed of downsampled processing
Solution Approach 2:
The initial disparity map obtained from downsampled images acts as an intermediary that bridges the gap between fast low-resolution computation and precise high-resolution computation. This intermediary disparity information guides the second-stage computation, allowing the system to achieve both speed and precision
Data Source
AI summary
A disparity determining method, an electronic device, and a computer-readable storage medium are provided. The disparity determining method includes: downsampling a first image and a second image, to obtain a first image with a first size and a second image with the first size; performing first-stage disparity computation on the first image with the first size and the second image with the first size, to obtain a first disparity map with the first size; and performing second-stage disparity computation on a first image with a second size and a second image with the second size based on an upsampled disparity map, with the second size, of the first disparity map, to obtain a second disparity map with the second size.