Wide Angle Image Disparity Estimation via Non-Linear Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wide angle cameras, such as those using fish eye lenses, introduce substantial distortion when capturing images, complicating disparity and depth estimation due to the interdependence of vertical and horizontal image positions, leading to inaccurate disparities and increased resource requirements for existing disparity estimation algorithms.
Innovation Solution
A method involving a receiver, mapper, and disparity estimator that applies non-linear vertical and horizontal mappings to a wide angle image to generate a modified image, allowing for efficient disparity estimation without requiring high-resolution conversions and enabling reuse of existing algorithms, while optimizing mapping functions for each direction independently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of moving object
If wide angle cameras with fish eye lenses are used to capture the scene, then the viewing angle and coverage are increased, but substantial distortion is introduced that complicates disparity and depth estimation
Solution Approach 1:
The patent introduces an intermediate projection step between the wide angle image and the disparity estimation process. The wide angle image is first converted to an intermediate projection that preserves the wide viewing angle while creating a more favorable geometry for disparity estimation, thereby mediating between the conflicting requirements of wide coverage and accurate measurement
Solution Approach 2:
The patent transforms the projection parameters of the wide angle image by converting from the original wide angle projection to an intermediate projection with different geometric properties. This parameter change modifies the relationship between image coordinates and scene coordinates, making disparity estimation more accurate while maintaining the wide viewing angle coverage
2Measurement precision
If existing disparity estimation algorithms are used on wide angle images, then disparity information can be obtained, but the algorithms become more complex and resource-intensive
Solution Approach 1:
The patent performs a preliminary transformation of the wide angle image to an intermediate projection before applying disparity estimation algorithms. This preliminary action prepares the data in a format that is more suitable for standard algorithms, reducing their complexity and computational requirements while maintaining measurement accuracy
3Measurement precision
If high-resolution conversion is applied to wide angle images for disparity estimation, then accuracy may be improved, but resource requirements increase substantially
Solution Approach 1:
The patent changes the projection parameters rather than increasing resolution, transforming the wide angle image to an intermediate projection that provides accurate disparity estimation at the original resolution. This parameter change approach achieves accuracy improvement without the substantial resource increase that would result from high-resolution conversion
Data Source
AI summary
An apparatus where a vertical image position of the scene point may depend on the horizontal image position. A mapper generates a modified image having a modified projection by applying a mapping to the first wide angle image corresponding to a mapping from the first projection to a perspective projection followed by a non-linear vertical mapping from the perspective projection to a modified vertical projection of the modified projection and a non-linear horizontal mapping from the perspective projection to a modified horizontal projection of the modified projection. Then a disparity estimator generates disparities for the modified image relative to a second image and representing a different view point than the first wide angle image.