Panoramic Video Stitching via Image Mode Homographies
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Camera arrays face challenges in achieving high spatial resolution for panoramic videos due to the reduction in spatial resolution caused by overlapping views during the video stitching process, as existing algorithms rely heavily on feature points in overlapping regions which may be limited, especially in areas with minimal features.
Innovation Solution
The method involves capturing images in addition to video frames by each camera, stitching these images first to determine image mode homographies, which are then used to derive video mode homographies, allowing for improved stitching and reduced overlapping regions, thereby increasing the spatial resolution of the final panoramic video.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If video stitching is performed using traditional feature point matching between video frames, then the stitching process can be completed, but the spatial resolution of the final panoramic video is reduced due to overlapping views
Solution Approach 1:
The patent performs image mode stitching before video stitching. By first stitching high-resolution images to obtain accurate homographies, then using these homographies for video stitching, the method eliminates the need for overlapping views in video mode, thereby preserving spatial resolution while maintaining stitching accuracy
Solution Approach 2:
The patent changes the operating mode from direct video mode stitching to image mode stitching followed by homography application. This parameter change (mode switching) allows the system to utilize higher resolution image captures for establishing transformations, which are then applied to video frames, resolving the spatial resolution degradation issue
2Manufacturing precision
If more cameras are used in the array to increase spatial resolution, then the resolution improves, but the cost and size of the array increases
Solution Approach 1:
The patent makes the image capture mode serve multiple functions: it provides both the source material for high-resolution stitching and the basis for calculating homographies needed for video stitching. This multi-functionality allows the system to achieve high spatial resolution without increasing the number of cameras, as the image mode captures serve dual purposes
3Reliability
If overlapping views are increased to ensure sufficient feature points for stitching, then stitching reliability improves, but the spatial resolution of the final panoramic video decreases
Solution Approach 1:
The patent performs stitching in image mode first, where overlapping views can be maximized to ensure reliable feature point matching and homography calculation. The resulting homographies are then applied to video frames without requiring overlapping views, thus achieving both high reliability in stitching and high spatial resolution in the final output
Data Source
AI summary
The present disclosure involves systems, software, and computer implemented methods for increasing spatial resolution of panoramic video captured by a camera array. In one example, a method may include identifying a captured image from each camera in a camera array associated with a capture of a panoramic video. The captured images are stitched together to generate at least one combined image and image mode homographies are calculated between the plurality of cameras in the camera array based on the stitching results. A plurality of captured video frames from each camera in the camera array are identified and video mode homographies of the plurality of cameras are determined based on the calculated image mode homographies. The determined video mode homographies are applied to stitch the plurality of captured video frames.