OmniTry: Virtual Try-On Anything without Masks

TL;DR

OmniTry introduces a unified, mask-free virtual try-on framework capable of handling diverse wearable objects beyond clothing, utilizing a two-stage training process with unpaired and paired images for improved localization and appearance consistency.

Contribution

The paper presents OmniTry, a novel unified VTON framework that extends to various wearable objects without masks, using a two-stage training pipeline with unpaired and paired data.

Findings

Outperforms existing methods in object localization and ID-preservation

Converges quickly with few paired samples in the first training stage

Achieves comprehensive results across 12 wearable object classes

Abstract

Virtual Try-ON (VTON) is a practical and widely-applied task, for which most of existing works focus on clothes. This paper presents OmniTry, a unified framework that extends VTON beyond garment to encompass any wearable objects, e.g., jewelries and accessories, with mask-free setting for more practical application. When extending to various types of objects, data curation is challenging for obtaining paired images, i.e., the object image and the corresponding try-on result. To tackle this problem, we propose a two-staged pipeline: For the first stage, we leverage large-scale unpaired images, i.e., portraits with any wearable items, to train the model for mask-free localization. Specifically, we repurpose the inpainting model to automatically draw objects in suitable positions given an empty mask. For the second stage, the model is further fine-tuned with paired images to transfer the consistency of object appearance. We observed that the model after the first stage shows quick convergence even with few paired samples. OmniTry is evaluated on a comprehensive benchmark consisting of 12 common classes of wearable objects, with both in-shop and in-the-wild images. Experimental results suggest that OmniTry shows better performance on both object localization and ID-preservation compared with existing methods. The code, model weights, and evaluation benchmark of OmniTry will be made publicly available at https://omnitry.github.io/.