Virtual 3D Image Reconstruction Using Native Disparity Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face challenges in reconstructing synthetic images from the perspective of a virtual camera based on images captured by multiple physical cameras, particularly in achieving accurate and efficient image reconstruction within the computational constraints of modern mobile devices and current communication architectures.
Innovation Solution
The method involves capturing images from multiple cameras, generating native disparity maps to solve the stereo correspondence problem, and using these maps along with calibration data to create synthetic images from a selected virtual camera position, employing techniques like rasterizing and ray-tracing to generate UV sample coordinates and sample weights for accurate image reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple physical cameras are used to capture images for virtual 3D reconstruction, then the quality and realism of the synthetic images are improved, but the computational complexity and processing cost increase significantly
Solution Approach 1:
The patent divides the complex image reconstruction task into separate processing stages: capturing images from multiple cameras, generating depth information through disparity maps, and synthesizing the final virtual view. This segmentation allows each stage to be optimized independently and processed efficiently on mobile devices.
Solution Approach 2:
The patent performs preliminary processing of camera images to generate disparity maps and depth information before the actual virtual view synthesis. By pre-computing these intermediate results, the system reduces the computational burden during real-time rendering and improves overall processing efficiency.
2Measurement precision
If high-quality synthetic images are generated through complex processing techniques, then the realism and accuracy of the virtual camera perspective is improved, but the processing time and computational resources required increase
Solution Approach 1:
The patent implements adaptive processing that adjusts the level of detail and computational effort based on the specific requirements of each scene region. Areas with less important visual information undergo simpler processing, while critical regions receive more intensive processing, optimizing the balance between quality and processing time.
3Measurement precision
If detailed disparity maps are generated to solve the stereo correspondence problem accurately, then the precision of depth information is improved, but the computational cost and memory requirements increase
Solution Approach 1:
The patent applies different levels of disparity map detail to different regions of the image based on their importance. Critical regions such as foreground objects and areas with significant depth variations receive high-resolution disparity information, while background regions use lower-resolution data, reducing overall computational load while maintaining perceived quality.
Data Source
AI summary
Methods, systems, devices and computer software/program code products enable reconstruction of synthetic images of a scene from the perspective of a virtual camera having a selected virtual camera position, based on images of the scene captured by a number of actual, physical cameras.