Real-Time Multiview Video Conversion Through Buffered Frame Blending
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video conversion techniques for emerging display technologies like holographic and 3D displays require significant pre-processing efforts to convert 2D video into multiview formats, limiting real-time rendering capabilities.
Innovation Solution
A method and system for converting 2D video streams into multiview video in real-time by using a computer-implemented process that involves loading 2D frames into a buffer, calculating multiview parameters, and dynamically blending time-adjacent frames based on camera baseline and center viewpoint, allowing for real-time rendering on multiview displays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If 2D video is converted to multiview format using traditional techniques, then the viewing experience on emerging displays is improved, but significant pre-processing time and computational resources are required
Solution Approach 1:
The patent extracts key viewpoint information from the 2D video stream in advance and stores it in a buffer, performing preliminary processing that enables rapid multiview conversion later without requiring extensive pre-processing of the entire video
Solution Approach 2:
The conversion process is divided into independent stages: extracting viewpoint data from individual frames, storing in buffer, and generating multiview outputs on-demand. This segmentation allows real-time processing rather than batch pre-processing
2Productivity
If multiview video is rendered in real-time from 2D video stream, then rendering speed is improved, but computational complexity increases
Solution Approach 1:
Viewpoint information is extracted and buffered in advance from the 2D stream, so that when multiview rendering is needed, the computational work has already been partially completed, reducing real-time complexity
Solution Approach 2:
Instead of performing complex 3D reconstruction, the patent creates simplified copy views by warping and blending the buffered 2D frames according to stored viewpoint parameters, reducing computational requirements
Data Source
AI summary
Systems and methods are directed to real-time multiview video conversion. This conversion may involve receiving a video stream including two-dimensional (2D) frames, where each 2D frame corresponds to a respective 2D video timestamp. In addition, a camera baseline and a center viewpoint are identified. These parameters may be user-specified or predetermined. A target timestamp for a view of a multiview frame may be determined based on the camera baseline and the center viewpoint. The view is generated from a subset of 2D frames having 2D video timestamps adjacent to the target timestamp. A multiview video is rendered for display, where the multiview video comprises the view of the multiview frame.