Virtual Fitting Image Processing for User-Orientation Mismatch
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional virtual fitting systems require users to stand facing the digital signage to confirm their fitting image, leading to uncomfortable experiences when compounding back half-body cloth images with front half-body user images.
Innovation Solution
An image processing apparatus that acquires and processes photographs to detect and alter the user region's luminance, allowing the calculation of the user's position and compounding a back cloth image with the altered image, enabling the generation of a fitting image without regard to the user's direction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the user stands facing the digital signage to confirm fitting image, then the fitting image can be displayed, but the cloth image direction does not match the user's actual body orientation causing discomfort
Solution Approach 1:
The patent inverts the conventional approach by not requiring the user to face the signage, instead processing the image to make the user's direction undecidable and compounding the cloth image accordingly. This allows the user to stand in any orientation while still achieving proper alignment between the cloth image and user body, resolving the contradiction between operational ease and alignment accuracy
Solution Approach 2:
The patent changes the luminance parameter of the user region in the photographing image to create an undecidable direction state. By modifying this visual parameter, the system can then compound the back cloth image with correct orientation regardless of the user's actual facing direction, thereby maintaining both user convenience and image alignment accuracy
2Device complexity
If the photographing image direction is used directly for compounding, then the processing is simple, but the fitting image becomes uncomfortable when user direction and cloth direction do not match
Solution Approach 1:
The patent performs preliminary action by changing the luminance of the user region before the compounding process. This preprocessing step makes the user's direction undecidable in advance, allowing the system to subsequently compound the back cloth image with correct orientation without requiring complex real-time direction analysis, thus balancing processing complexity with fitting image comfort
Solution Approach 2:
By changing the luminance parameter of the user region, the patent creates a transformed image state that enables proper compounding of back cloth images. This parameter modification approach maintains relatively simple processing while significantly improving fitting image comfort by ensuring correct directional alignment
Data Source
AI summary
According to one embodiment, a photographing image including a user region is acquired. The user region is detected from the photographing image. The user region in the photographing image is changed so that a user's direction in the user region cannot be decided, and the photographing image in which the user region is changed is acquired as a change image. A position of the user region in the photographing image is calculated. An image of a back side-cloth is compounded with the user region in the change image at the position.