Immersive Video Processing Atlas Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current immersive video technologies face challenges in efficiently supporting motion parallax for rotational and translation motions, particularly in reducing data size and metadata transmission while maintaining high-quality 6 DoF video reproduction.
Innovation Solution
The method involves determining a priority order for pruning source view videos, generating residual videos, packing patches into an atlas video, and encoding it, with the basic view video having the highest priority and additional view videos prioritized based on overlapping area size and region of interest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all source view videos are transmitted to support motion parallax, then video quality and immersion are improved, but data transmission size increases
Solution Approach 1:
The patent extracts and removes duplicate video data from multiple source views by identifying overlapping regions and retaining only unique content. This extraction process eliminates redundant information while preserving all necessary visual data for 6DoF motion parallax reproduction, thereby reducing overall data size without compromising video quality.
Solution Approach 2:
The patent merges multiple source view videos into a single consolidated video file by integrating unique regions from each view. overlapping areas are detected and consolidated, with duplicate content removed and unique content from each view combined into a unified structure that supports motion parallax while minimizing total data volume.
2Adaptability or versatility
If multiple additional view videos are kept to support translation motion, then motion parallax capability is improved, but processing complexity increases
Solution Approach 1:
The patent performs preliminary processing to detect and mark duplicate regions across multiple source views before final video generation. By identifying overlapping areas in advance and establishing a consolidation strategy, the system prepares the data structure to efficiently support translation motion without requiring complex real-time processing during playback.
Solution Approach 2:
The patent changes the organizational parameters of video data by transforming multiple separate additional view videos into a consolidated structure with optimized region allocation. This parameter change reorganizes how video content is stored and accessed, reducing processing complexity while maintaining full motion parallax capability for translation movements.
3Productivity
If duplicate data among additional view videos is removed, then data transmission efficiency is improved, but data structure complexity increases
Solution Approach 1:
The patent segments the video data into distinct regions identified by bounding boxes, where each segment represents a unique non-duplicate area from a specific source view. This segmentation approach organizes removed duplicate data into manageable, clearly defined segments that can be efficiently stored and transmitted, balancing structure complexity with transmission efficiency.
Data Source
AI summary
An immersive video processing method according to the present disclosure includes determining a priority order of pruning for source view videos, generating a residual video for an additional view video based on the priority order of pruning, packing a patch generated based on the residual video into an atlas video, and encoding the atlas video.