Virtual Try-On Components Using Pre-Rendered Images to Reduce Lag
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing VTO engines face delays and privacy concerns due to remote execution, which reduces the real-time simulation experience and requires significant communication lag, while local processing involves engine download delays.
Innovation Solution
Implementing pre-rendered 'after' model images using a VTO engine on a user device, allowing simultaneous presentation of model and user image experiences, with the model experience being presented while the user image experience is being readied, thus reducing lag and enhancing privacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If VTO engine is executed remotely in the cloud, then communication lag is reduced and processing power is improved, but significant communication lag occurs and privacy is compromised
Solution Approach 1:
The patent pre-renders model images before the user interacts with the system. These pre-rendered images are stored and can be immediately displayed when the user opens the application, eliminating the need to wait for real-time processing during the initial interaction phase.
2Reliability
If VTO engine is executed locally on user device, then privacy is enhanced and real time simulation is improved, but engine download delays occur
Solution Approach 1:
The system pre-renders model images before the user needs to interact with the VTO functionality. This preliminary rendering occurs during server-side preparation, so when the user accesses the application, the images are already ready for immediate display, eliminating download and processing delays.
Solution Approach 2:
Instead of downloading the entire VTO engine to the user's device, the system uses pre-rendered copies of the processed images. These static or pre-processed image copies can be displayed immediately without requiring the user to download and initialize the full engine, significantly reducing startup time while maintaining privacy through local execution.
3Loss of time
If pre-rendered model images are presented while user image experience is readied, then lag is reduced and real time experience is improved, but model images must be prepared in advance
Solution Approach 1:
The system performs model image rendering in advance during server-side preparation before the user interacts with the application. These pre-rendered images are stored and can be immediately displayed when the user opens the application, eliminating the need to wait for real-time processing during the initial interaction phase.
Data Source
AI summary
Methods and systems for VTO provide an augmented reality experience. Webpage components define different VTO experiences for different products/services and different instances (e.g., hair, nail or makeup products with different colors). The components can integrate with a webpage, for example, as an advertisement. When executed on a user device, the components provide a VTO for a particular product including a collection of interfaces that provide a model image experience. Pre-rendered “after” model images using an appropriate VTO engine show the product/services. The after model images can be presented prior to a live user image experience enabled by a corresponding user device executed VTO engine rendering a product instance to a user image. The model experience can be presented while the user image experience (e.g. its VTO engine) is readied. The model experience can be presented in a same/overlapping region as the user image experience, for example.