Image Stitching with Audio Synchronization and FOV Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image stitching methods for spherical panorama images captured by fisheye lenses lack flexibility in adjusting the field of view (FOV) and image effects, making it difficult for users to achieve optimal stitching results.
Innovation Solution
An image stitching method and device that receive video files from multiple fisheye lenses with different FOVs, synchronize audio tracks to adjust image timing, convert images into adjusted formats, and perform image stitching to generate a stitched image, allowing for flexible FOV and image effect adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional image stitching methods are used with fixed stitching points, then the stitching process is simple, but the field of view (FOV) and image effects cannot be freely adjusted
Solution Approach 1:
The patent implements dynamic stitching point selection and adjustable FOV parameters, allowing the stitching process to adapt to different user requirements. The system enables real-time adjustment of stitching positions and image effects rather than using fixed predetermined parameters, directly resolving the contradiction between adaptability and complexity.
Solution Approach 2:
The patent allows users to modify multiple parameters including stitching points, FOV angles, and image processing effects. By enabling parameter adjustment, the system achieves versatility in FOV and image effect control while maintaining a manageable complexity level through standardized parameter interfaces.
2Measurement precision
If audio tracks from multiple video files are synchronized, then image timing accuracy is improved, but additional processing time is required
Solution Approach 1:
The system performs automatic audio-based synchronization without requiring manual intervention. The audio tracks self-correct timing deviations through automated correlation algorithms, achieving high timing precision while minimizing the need for user time investment in the synchronization process.
Solution Approach 2:
The patent performs audio synchronization as a preliminary step before image stitching. By pre-synchronizing the audio tracks and corresponding images, the system establishes accurate timing relationships in advance, which then streamlines the subsequent stitching process and reduces overall processing time.
3Adaptability or versatility
If images from lenses with different FOVs are stitched, then flexibility in image composition is improved, but alignment difficulty increases
Solution Approach 1:
The patent introduces intermediate processing steps including FOV normalization and coordinate transformation as mediators between images from different lenses. These intermediary processes standardize the different FOVs and facilitate accurate alignment, resolving the contradiction between compositional flexibility and alignment difficulty.
Solution Approach 2:
The system handles different FOVs by transforming images into a unified coordinate system and applying multi-dimensional adjustment parameters. This dimensional transformation approach enables flexible composition of images from lenses with different FOVs while maintaining precise alignment through mathematical coordinate mappings.
Data Source
AI summary
An image stitching method for use in an image processing apparatus is provided. The method includes the steps of: receiving a first video file and a second video file from an image capturing unit, wherein the first video file includes at least one first image and a first audio track, and the second video file includes at least one second image and a second audio track; calculating delay time between a first acoustic feature of the first audio track and a second acoustic feature of the second audio track to synchronize the first image and the second image; converting the first image and the second image into a first adjusted image and a second adjusted image; and performing an image stitching process on the first adjusted image and the second adjusted image to generate a stitched image.