Distributed Virtual Viewpoint Image Generation Reducing Network Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for generating virtual viewpoint images from multiple camera viewpoints face challenges in managing increased network transmission loads and server arithmetic loads as the number of cameras increases.
Innovation Solution
A distributed image processing system where multiple cameras and microphones capture images and sound, with a network of image processing apparatuses generating virtual viewpoint images by processing and combining information from multiple camera viewpoints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple cameras are connected through a centralized server to generate virtual viewpoint content, then the system can collect and process images from multiple viewpoints, but the network transmission load and server arithmetic load increase in accordance with the number of cameras
Solution Approach 1:
The patent divides the centralized server system into multiple distributed image processing apparatuses, each handling a subset of cameras. This segmentation reduces the load on any single apparatus and distributes network traffic, thereby reducing overall network transmission load while maintaining multi-viewpoint content generation capability.
Solution Approach 2:
The patent introduces a hierarchical processing architecture where image processing occurs at multiple levels (local processing apparatuses and higher-level synthesis apparatuses). This dimensional change in processing organization allows load distribution across the network hierarchy, reducing transmission loads on individual network segments.
2Adaptability or versatility
If multiple cameras are connected through a centralized server to generate virtual viewpoint content, then the system can collect and process images from multiple viewpoints, but the server arithmetic load increases in accordance with the number of cameras
Solution Approach 1:
The patent segments the arithmetic processing tasks across multiple independent image processing apparatuses. Each apparatus processes images from a subset of cameras locally, performing 3D model generation and rendering tasks distributedly. This segmentation reduces the arithmetic load on any single server while maintaining overall processing capability for multi-viewpoint content.
Solution Approach 2:
The patent introduces intermediate processing nodes (image processing apparatuses) between cameras and the final output system. These intermediaries perform preliminary processing (image collection, 3D model generation) locally, reducing the computational burden on central servers and enabling scalable multi-viewpoint content generation.
3Productivity
If images are collected and processed in a centralized system, then virtual viewpoint content can be generated, but the system complexity increases with the number of cameras
Solution Approach 1:
The patent segments the centralized processing system into multiple modular image processing apparatuses, each handling a subset of cameras independently. This modular segmentation reduces system complexity by creating manageable, repeatable units that can be scaled without proportionally increasing overall system complexity.
Solution Approach 2:
Each image processing apparatus in the patent performs self-contained processing of images from its assigned cameras, including local 3D model generation and rendering. This self-service capability reduces inter-dependency between system components, simplifying system management and reducing overall complexity while maintaining high productivity.
Data Source
AI summary
A method for generating a virtual viewpoint image includes generating, by a first image processing apparatus, first information for generating the virtual viewpoint image based on a first image captured by a first camera, generating, by a second image processing apparatus, second information for generating the virtual viewpoint image based on a second image captured by a second camera; specifying a viewpoint in the virtual viewpoint image; and generating, by a third image processing apparatus, a virtual viewpoint image corresponding to the specified viewpoint using the first information and the second information.