Vertical Disparity Correction in Stereoscopic Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Stereoscopic image pairs often contain vertical disparities due to lens distortion and camera misalignment, leading to an unenjoyable viewing experience, as conventional methods for correcting these disparities are inaccurate, redundant, and computationally intensive.
Innovation Solution
A method and apparatus using a deep neural network to generate an optical flow map, extract vertical motion vectors, and vertically align stereoscopic images based on determined disparities, eliminating the need for preprocessing and enabling real-time correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods are used to correct vertical disparities, then some level of correction is achieved, but the process becomes computationally intensive and redundant
Solution Approach 1:
The patent extracts only the essential vertical motion vector information from the optical flow map, discarding redundant horizontal and other motion components. This extraction focuses computational resources on the specific parameter needed for vertical disparity correction, achieving both accuracy and efficiency by eliminating unnecessary processing.
Solution Approach 2:
The patent transforms the optical flow data into a simplified representation focusing on vertical motion parameters. By changing the parameter representation from full optical flow vectors to extracted vertical motion values, the system achieves precise vertical disparity measurement without the computational burden of processing complete flow fields.
2Measurement precision
If preprocessing steps are added to correct vertical disparities, then correction accuracy improves, but the complexity and time consumption increase
Solution Approach 1:
The patent performs preliminary extraction of vertical motion vectors from the optical flow map during the initial processing stage. By preparing this specific information upfront in a simplified form, the subsequent vertical alignment operation becomes straightforward and efficient, avoiding the need for complex preprocessing pipelines while maintaining high accuracy.
3Measurement precision
If pixel-to-pixel level correction is implemented, then vertical disparity precision is maximized, but processing time increases
Solution Approach 1:
The patent extracts vertical motion vector values directly from the optical flow map at each pixel location, obtaining precise vertical disparity information without requiring complex pixel-to-pixel comparison operations. This extraction approach achieves maximum precision with minimal processing time by focusing only on the vertical component.
Solution Approach 2:
The patent replaces traditional mechanical image comparison methods with a data-driven approach using extracted motion vector values. Instead of performing computationally intensive pixel matching, the system uses the extracted vertical motion information to directly determine and correct disparities, achieving both precision and speed.
Data Source
AI summary
Due to the factors such as lens distortion and camera misalignment, stereoscopic image pairs often contain vertical disparities. Introduced herein is a method and apparatus that determine and correct vertical disparities in stereoscopic image pairs using an optical flow map. Instead of discarding vertical motion vectors of the optical flow map, the introduced concept extracts and analyzes the vertical motion vectors from the optical flow map and vertically aligns the images using the vertical disparity determined from the vertical motion vectors. The introduced concept recognizes that although not apparent, vertical motion does exist in stereoscopic images and can be used to correct the vertical disparity in stereoscopic images.