Virtual Object Outfit Transfer via Image Segmentation and Region Filling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for dressing up virtual objects in virtual scenes require cumbersome three-dimensional modeling and cloth simulation, leading to low wearing efficiency and increased processing resources.
Innovation Solution
A data processing method that segments an outfit image from an image of a real object, obtains a first image of a virtual object driven by a three-dimensional posture parameter of the real object, and fills the wearing region in the virtual object's image with the outfit image, reducing the need for three-dimensional modeling and simulation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If three-dimensional modeling and cloth simulation are performed on garments for virtual objects, then the physical movement state of clothes (wrinkles, swings) is restored, but the data processing complexity and time consumption increase significantly
Solution Approach 1:
The patent uses image copying technology to transfer the outfit from the real object directly to the virtual object. Instead of creating three-dimensional models and performing complex cloth simulations, the system captures images of the real object wearing the outfit, segments the outfit from the background, and copies it onto the virtual object. This approach maintains visual realism while dramatically reducing data processing complexity and time consumption.
2Manufacturing precision
If three-dimensional modeling and cloth simulation are performed on garments for virtual objects, then the physical movement state of clothes (wrinkles, swings) is restored, but the processing time increases
Solution Approach 1:
The system performs rapid outfit transfer by copying segmented outfit images directly onto virtual objects. This eliminates the time-consuming processes of three-dimensional modeling and frame-by-frame cloth simulation, reducing the outfit change process from potentially minutes to seconds while maintaining visual quality through image copying and compositing techniques.
3Manufacturing precision
If three-dimensional modeling and cloth simulation are performed on garments for virtual objects, then the physical movement state of clothes (wrinkles, swings) is restored, but the computational resources required increase
Solution Approach 1:
The patent replaces computationally intensive three-dimensional modeling and cloth simulation with efficient image copying operations. By segmenting the outfit from the real object and copying it onto the virtual object, the system achieves similar visual results with significantly reduced computational resource consumption, making real-time outfit changes feasible.
4Productivity
If image segmentation and filling methods are used to dress virtual objects, then data processing efficiency improves, but the complexity of image processing algorithms increases
Solution Approach 1:
The patent applies image segmentation to separate the outfit from the real object in captured images. This segmentation process divides the complex task of outfit transfer into manageable steps: segmenting the outfit region, extracting the outfit image, and filling it onto the virtual object. This structured approach improves efficiency while keeping each processing step relatively simple and well-defined.
Data Source
AI summary
This application disclose a method for changing an outfit of a virtual object in a virtual scene performed by a computer device. The method includes: segmenting an outfit image of a target outfit from an image of a real object wearing the target outfit; obtaining a first image of the virtual object from a first video of the virtual object, the first video including an action of the virtual object driven by a three-dimensional posture parameter of the real object in a real world; and filling a wearing region in the first image of the virtual object with the outfit image, to obtain an image of the virtual object wearing the target outfit in the virtual scene.