3D Subject Image Editing from Single Capture Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing apparatuses, such as those described in Japanese Patent Laid-Open No. 2022-076733, are limited to 2D image processing, requiring repeated shots or re-takes to achieve desired poses and expressions in captured images, which is inefficient for generating desired outputs.
Innovation Solution
An image processing apparatus that generates 3D data based on captured images and asperity information, allowing for changes to the form of the 3D data to produce desired images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If 2D image processing is used, then the processing method is simple, but the ability to capture desired poses and expressions is limited requiring repeated shots
Solution Approach 1:
The patent transforms 2D captured images into 3D data models, enabling manipulation of the subject from multiple angles and perspectives. This dimensional transition allows users to adjust poses and expressions without retaking shots, directly resolving the contradiction between processing simplicity and output generation efficiency.
Solution Approach 2:
The system performs preliminary 3D data generation and storage before the actual image output is needed. By creating a 3D model from the captured image in advance, the system enables rapid generation of various poses and expressions without requiring repeated shooting, thus improving productivity while maintaining processing efficiency.
2Manufacturing precision
If repeated shots are taken to achieve desired poses, then the desired image can be obtained, but the time and resources are wasted
Solution Approach 1:
By converting 2D images to 3D data, the system enables retrieval of desired poses and expressions from the 3D model without requiring multiple shooting attempts. This eliminates time loss while maintaining the ability to achieve high-quality images with correct poses and expressions.
Solution Approach 2:
The system creates a 3D copy or model of the subject from a single captured image. This digital copy can then be manipulated to generate various poses and expressions without needing to recapture the subject, saving time while maintaining image quality.
3Device complexity
If conventional 2D processing is used, then the device complexity is low, but the versatility of image manipulation is limited
Solution Approach 1:
The patent introduces 3D data structures and processing capabilities to enable versatile manipulation of subject images. While this increases device complexity, it provides the adaptability to generate images with various poses, expressions, and angles from a single capture, significantly improving manipulation flexibility.
Data Source
AI summary
An image processing apparatus generates an image pertaining to a subject. The apparatus obtains a captured image and asperity information of the subject; generates 3D data of the subject based on the captured image and the asperity information obtained; makes a change instruction to change a form of the 3D data generated; and generates, as the image pertaining to the subject, an image corresponding to the 3D data changed based on the change instruction.