Electronic Device Image Synthesis for Virtual Try-On
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face challenges in visualizing how clothing or accessories would fit and match their personal style and environment when purchasing online, as they cannot physically try on items and adapt to changing seasons and environments.
Innovation Solution
An electronic apparatus that selects and synthesizes user pictures based on weighted values, attribute information, and context to combine with product images, allowing users to visualize themselves wearing the products.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If users purchase clothes or accessories online by looking at product images worn by models or avatars, then they can make purchasing decisions, but they cannot visualize how the actual products would fit and match their personal style and environment
Solution Approach 1:
The patent creates a virtual copy of the user by synthesizing their actual photograph with product images. The system overlays the clothing product onto the user's image, generating a composite image that shows how the user would appear wearing the product. This copying approach allows users to visualize the actual fit and appearance without physically trying on the clothes, thereby recovering the lost visualization information while maintaining online purchasing convenience.
Solution Approach 2:
The patent introduces an intermediary system that acts as a bridge between the user and the product. This intermediary (the image synthesis system) processes both the user's photograph and the product image to generate a composite visualization. The intermediary translates the abstract product image into a context-specific visualization that shows how the product would appear on the specific user in their specific environment, solving the information loss problem.
2Adaptability or versatility
If users cannot try on clothes physically, then online purchasing is convenient, but users cannot adapt to changing seasons and environments
Solution Approach 1:
The patent performs preliminary visualization before the user makes a purchasing decision. By generating the composite image showing how the user would appear in different clothing options, the system allows users to evaluate suitability for their specific environment and season in advance. This preliminary action enables users to make more informed decisions about whether the clothing is appropriate for their current environmental context, thereby improving adaptability without extending decision-making time.
Solution Approach 2:
The patent enables dynamic adaptation by allowing users to generate composite images for different products, seasons, and environments on demand. The system can dynamically create multiple visualization scenarios, showing how the user would appear in different clothing options suited for different environmental conditions. This dynamic capability allows users to adapt their purchasing decisions to changing seasons and environments without losing time on physical trial sessions.
3Measurement precision
If users rely on product images with models or avatars, then purchasing decisions can be made, but the representation does not accurately reflect how the product would appear on the actual user
Solution Approach 1:
The patent applies local quality by processing the user's image to extract specific local characteristics such as body shape, size, and features. The image synthesis system selectively adapts the product image to match these local qualities of the user, ensuring that the clothing fits and appears realistically on the user's specific body type. This local customization approach significantly improves fit accuracy compared to generic model images, while the processing complexity is managed through automated algorithms that focus on key local features.
Data Source
AI summary
An apparatus and method for synthesizing an image and an electronic apparatus using the method are provided. The electronic apparatus includes a memory that stores pictures, a transceiver that communicates with an external server, a display that detects a user interaction, and a processor that reads pictures from the memory. The processor first selects pictures based on a weighted value, determines first attribute information and second attribute information from the first selected pictures, stores the first attribute information and the second attribute information in a database form, second selects at least one picture from among the first selected pictures based on the first attribute information, the second attribute information and user context information, controls the transceiver to receive a product image corresponding to the second selected picture from the external server, and creates a composite image by synthesizing the second selected picture and the product image.