3D Model Replacement for Faster Virtual Video Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for reconstructing video images from captured images of a target object are inefficient, limiting the effective utilization of three-dimensional model data for virtual space applications.
Innovation Solution
An image processing method that generates first three-dimensional model data from multiple images of a model's exterior and replaces second three-dimensional model data with template data to reconstruct a video image, utilizing a processing unit and output unit to facilitate efficient video image reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If primitive mapping technique is used to generate virtual space images from captured images, then virtual space reconstruction is possible, but reconstruction processing efficiency is low
Solution Approach 1:
The patent applies preliminary action by pre-generating three-dimensional model data from captured images before the actual video reconstruction process. The system creates a reusable three-dimensional model data structure that can be efficiently referenced during video reconstruction, eliminating the need to process raw images repeatedly for each video frame.
Solution Approach 2:
The patent uses copying by creating a three-dimensional model data representation that serves as a digital copy of the captured object. This three-dimensional model can be efficiently manipulated and rendered for video reconstruction without requiring access to or processing of the original captured images, significantly improving processing efficiency.
2Manufacturing precision
If three-dimensional model data is generated and integrated into virtual spaces, then video image quality is improved, but processing complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the video reconstruction process into distinct stages: image capture, three-dimensional model generation, and video reconstruction. Each stage processes specific data types and produces defined outputs, making the overall complex process more manageable and efficient.
Solution Approach 2:
The patent introduces three-dimensional model data as an intermediary between captured images and final video output. This intermediary representation simplifies the reconstruction process by providing a structured format that is easier to manipulate than raw images, reducing processing complexity while maintaining quality.
Data Source
AI summary
An image processing method of reconstructing a video image includes: causing a processor to generate, from a plurality of images captured of an exterior of a model in a plurality of directions and generated by an imaging device, first three-dimensional model data of the model included in the images; causing the processor to reconstruct the video image by replacing second three-dimensional model data with the first three-dimensional model data in template data of a video image constructed in advance to include the second three-dimensional model data; and causing an output device to output information indicating a destination from which the reconstructed video image is acquired.