Disparity Estimation Using Dual Cameras with Different Fields of View
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing systems struggle to accurately estimate real-world depth using cameras with different fields of view, limiting applications such as Bokeh effect processing, 3D object reconstruction, and virtual reality, especially with non-professional cameras or those with smaller lenses.
Innovation Solution
A method and system utilizing two cameras with different fields of view, one with a wide FOV and the other with a narrower FOV, to generate a disparity estimate by capturing overlapping and union FOV images, merging these estimates using deep neural networks for enhanced depth estimation across the entire FOV.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If cameras with different fields of view are used, then depth estimation accuracy is improved, but system complexity increases
Solution Approach 1:
The patent segments the field of view into multiple regions (first FOV region from wide-angle camera, second FOV region from telephoto camera) and processes each region separately with appropriate disparity estimation methods, then merges the results to achieve accurate depth estimation across the entire scene
Solution Approach 2:
The patent combines images and disparity estimates from multiple cameras with different FOVs by merging the first image from the wide-angle camera with the second image from the telephoto camera, and merging their respective disparity estimates to produce a comprehensive depth map
2Area of stationary object
If multiple cameras with different FOVs are used, then coverage area is improved, but processing complexity increases
Solution Approach 1:
The patent applies different processing approaches to different regions: the wide-angle camera processes the broader scene with lower detail requirements, while the telephoto camera focuses on the central region with higher detail requirements, optimizing processing complexity for each local area
3Device complexity
If disparity estimation is performed on overlapping FOV only, then processing simplicity is maintained, but depth estimation accuracy deteriorates
Solution Approach 1:
The patent extends disparity estimation from the traditional overlapping FOV region into the non-overlapping regions by using the first image from the wide-angle camera to provide depth information for areas not captured by the telephoto camera, effectively adding spatial dimension coverage
Data Source
AI summary
An electronic device and method are herein disclosed. The electronic device includes a first camera with a first field of view (FOV), a second camera with a second FOV that is narrower than the first FOV, and a processor configured to capture a first image with the first camera, the first image having a union FOV, capture a second image with the second camera, determine an overlapping FOV between the first image and the second image, generate a disparity estimate based on the overlapping FOV, generate a union FOV disparity estimate, and merge the union FOV disparity estimate with the overlapping FOV disparity estimate.