Related Experiment Video
Updated: Sep 21, 2025
09:17
Using a Virtual Store As a Research Tool to Investigate Consumer In-store Behavior
Published on: July 24, 2017
11.5K
Designing an AI-Based Virtual Try-On Web Application
Davide Marelli1, Simone Bianco1, Gianluigi Ciocca1
1Department of Informatics, Systems and Communication, University of Milano-Bicocca, Viale Sarca 336, 20126 Milano, Italy.
Sensors (Basel, Switzerland)
|May 28, 2022
Summary
This study introduces an advanced augmented reality eyewear virtual try-on system using deep learning and 3D face reconstruction. The markerless application provides realistic fit estimation and achieved high user satisfaction in usability testing.
Area of Science:
- Computer Vision
- Augmented Reality
- Artificial Intelligence
Background:
- Augmented Reality (AR), Virtual Reality (VR), and Artificial Intelligence (AI) are increasingly used across various sectors.
- Virtual try-on applications in retail allow users to visualize accessories, makeup, hairstyles, hair color, and clothing on themselves.
- Eyewear try-on is a significant application within the virtual try-on market.
Purpose of the Study:
- To propose and develop an eyewear virtual try-on experience using advanced deep learning and computer vision.
- To create a system capable of realistic virtual try-on on a 3D face model reconstructed from a single image.
- To ensure accurate fit estimation by considering actual eyewear and face dimensions in a markerless approach.
Main Methods:
- Leveraging deep learning-based computer vision techniques for virtual try-on.
- Reconstructing a 3D face model from a single input image.
- Assessing and comparing various face reconstruction approaches based on design, complexity, geometry errors, and texture quality.
- Developing a web application for user interaction, optimized for desktop and mobile devices.
Main Results:
- Selection of the most suitable face reconstruction approach for the virtual try-on framework through extensive analysis and experimentation.
- Implementation of a markerless system that accurately estimates eyewear fit based on real-world measurements.
- Development of an interactive web application for a seamless user experience.
- A usability study indicated that the eyewear virtual try-on application scored above average.
Conclusions:
- The proposed framework effectively integrates deep learning and computer vision for an advanced eyewear virtual try-on experience.
- The system offers realistic virtual try-on with accurate fit estimation, enhancing the online retail experience for eyewear.
- The markerless approach and user-friendly web application contribute to high usability and user satisfaction.
Related Concept Videos
01:29Impression Management Techniques I: Managing Appearances
6
Appearance is a multidimensional aspect of self-presentation that encompasses observable attributes such as clothing, grooming, speech, and nonverbal behavior. These elements are often strategically managed to align with socially constructed expectations in different settings. For instance, individuals tailor their appearance during job interviews, social gatherings, or athletic events to meet the perceived norms of those environments.Contextual Adaptation and Social SignalsThe research...
6
01:12Modeling and Similitude
350
Scaled modeling is a fundamental technique in engineering, enabling the study of large and complex systems by creating smaller, manageable replicas that recreate critical characteristics of the original. In hydrology and civil infrastructure, for example, scaled models of dams help analyze water flow, turbulence, and pressure. This method allows for accurate predictions of real-world behavior within a controlled environment, significantly reducing the cost and time involved in full-scale...
350