Virtual Garment Rendering With Illumination Estimation for Realistic Try-On
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The mismatch between real light source information and manually configured light source information in virtual try-on technologies leads to issues such as confusion in light source position and uneven illumination, affecting the realism of the try-on effect image.
Innovation Solution
An image generation method that includes acquiring a first image of a target object, performing illumination estimation to determine light information, rendering a target garment model based on this information, and fusing the images to create a realistic effect image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manually configured light source information is used in virtual try-on, then the process is simple and fast, but the light source position becomes confused and illumination becomes uneven
Solution Approach 1:
The patent introduces an intermediary component (light source position determination module) that bridges the gap between simple manual configuration and accurate illumination. This module automatically determines light source positions based on the input image, serving as a mediator that enhances illumination accuracy without significantly complicating the overall process.
Solution Approach 2:
The system performs self-service by automatically analyzing the input image to determine light source positions and illumination parameters. Instead of relying entirely on manual configuration, the system uses image processing algorithms to self-determine the appropriate lighting conditions, thereby improving accuracy while maintaining efficiency.
2Ease of operation
If manually configured light source information is used in virtual try-on, then the configuration process is simple, but the light source position becomes confused
Solution Approach 1:
The patent introduces an intermediary component (light source position determination module) that bridges the gap between simple manual configuration and accurate illumination. This module automatically determines light source positions based on the input image, serving as a mediator that enhances illumination accuracy without significantly complicating the overall process.
Solution Approach 2:
The patent replaces the mechanical/manual configuration process with an automated image processing system. Instead of manually specifying light source positions, the system uses computational algorithms to automatically determine positions from the input image, substituting manual operation with automated processing while maintaining ease of use.
3Ease of operation
If manually configured light source information is used in virtual try-on, then the setup is straightforward, but uneven illumination occurs
Solution Approach 1:
The system performs self-service by automatically analyzing the input image to determine light source positions and illumination parameters. Instead of relying entirely on manual configuration, the system uses image processing algorithms to self-determine the appropriate lighting conditions, thereby improving accuracy while maintaining efficiency.
Solution Approach 2:
The patent introduces an intermediary component (illumination determination module) that automatically adjusts illumination parameters based on image analysis. This intermediary ensures uniform and realistic lighting by computing appropriate light distribution, maintaining simplicity while eliminating illumination unevenness.
Data Source
AI summary
The embodiments of the present disclosure relate to the technical field of image processing. Provided are an image generation method and apparatus. The method comprises: acquiring a first image that comprises a target object; performing illumination estimation on the first image so as to acquire illumination information of the first image; rendering a target garment model according to the illumination information so as to generate a second image; and fusing the first image with the second image to acquire an effect image after the target object wears a virtual garment corresponding to the target garment model.


