Visual Modeling Challenges for 2D-3D Virtual Try-On

CVPR 2025 Workshop, Nashville TN

June 12, 2025


teaser

Introduction

Virtual Try-On (VTO) is an emerging consumer application that enables users to perceive products on their individual bodies in a virtual or mixed reality space. This has been driven by advances in 3D modeling of humans and objects (such as furniture, shoes, garments etc.). These experiences enable users to visualize and “feel” products, especially in the beauty, fashion, and accessories space, virtually, before making purchases. This gives buyers opportunities to customize and personalize products. Successful integration of realistic try-on experiences on e-commerce websites can have significant environmental impact by reducing the need to return products, improving satisfaction of purchased items, and improving accessibility. Beyond fashion and beauty, adoption of try-on-like experiences, especially in an Augmented Reality (AR) setting, can be seen in applications such as remote collaboration, gaming, education, and health-care. Enabling such applications, however, is not without its unique challenges that necessitate solutions inspired by ideas in computer vision, 3D modeling and reconstruction (single and multi-view), geometry processing, and more recently, generative AI. Our workshop will focus on these key technical areas of strong interest for the CVPR community. This is an active and multi-disciplinary area of research, drawing interest from researchers in academia and industry alike. Furthermore, the user-centric nature of this domain and inherent application to fashion requires researchers and technologists to understand the nature of bias in their data (catalog data is skewed towards a narrow BMI-range and demographic), trust and data-security issues (body image / data is often a necessary input) and the mental-health impact of try-on products for their users.


Invited Speakers

Ira Kemelmacher-Shlizerman
University of Washington
Michael Black
MPI Perceiving Systems
Gordon Wetzstein
Stanford University
Christian Theobalt
MPI Informatics
Sunil Hadap
Amazon Science
Huamin Wang
Style3D


Schedule

Welcome and Introduction 1:00 - 1:10 PM
Invited Talk 1 (Gordon Wetztein) 1:10 - 1:35 PM
Invited Talk 2 1:35 - 2:00 PM
Invited Talk 3 2:00 - 2:25 PM
Invited Talk 4 2:25 - 2:50 PM
Break 2:50 - 3:00 PM
Invited Talk 5 3:00 - 3:25 PM
Invited Talk 6 3:25 - 3:50 PM
Invited Talk 7 3:50 - 4:15 PM
Invited Talk 8 (Ira Kemelmacher-Schlizerman) 4:15 - 4:40 PM
Panel Discussion and Conclusion 4:45 - 5:30 PM




Organizers

Akshay Gadi Patil (POC)
Amazon Science
Vidya Narayanan
Amazon Science
Haoye Dong
Carnegie Mellon University
Gerard Pons-Moll
University of Tubingen
Ming C Lin
University of Maryland and Amazon Science