English

TryOnDiffusion: A Tale of Two UNets

Computer Vision and Pattern Recognition 2023-06-16 v1 Graphics

Abstract

Given two images depicting a person and a garment worn by another person, our goal is to generate a visualization of how the garment might look on the input person. A key challenge is to synthesize a photorealistic detail-preserving visualization of the garment, while warping the garment to accommodate a significant body pose and shape change across the subjects. Previous methods either focus on garment detail preservation without effective pose and shape variation, or allow try-on with the desired shape and pose but lack garment details. In this paper, we propose a diffusion-based architecture that unifies two UNets (referred to as Parallel-UNet), which allows us to preserve garment details and warp the garment for significant pose and body change in a single network. The key ideas behind Parallel-UNet include: 1) garment is warped implicitly via a cross attention mechanism, 2) garment warp and person blend happen as part of a unified process as opposed to a sequence of two separate tasks. Experimental results indicate that TryOnDiffusion achieves state-of-the-art performance both qualitatively and quantitatively.

Keywords

Cite

@article{arxiv.2306.08276,
  title  = {TryOnDiffusion: A Tale of Two UNets},
  author = {Luyang Zhu and Dawei Yang and Tyler Zhu and Fitsum Reda and William Chan and Chitwan Saharia and Mohammad Norouzi and Ira Kemelmacher-Shlizerman},
  journal= {arXiv preprint arXiv:2306.08276},
  year   = {2023}
}

Comments

CVPR 2023. Project page: https://tryondiffusion.github.io/

R2 v1 2026-06-28T11:04:40.852Z