Academic Project Page

Zhenchen Wan1, Yanwu Xu2, Dongting Hu1, Weilun Cheng1, Tianxi Chen1, Zhaoqing Wang2 Feng Liu1 Tongliang Liu2 Mingming Gong1,3
1University of Melbourne, Australia
2University of Sydney, Australia
3Mohamed bin Zayed University of Artificial Intelligence, UAE
Paper Code arXiv
TED-VITON Teaser

We propose a Mask-Free Virtual Try-On framework that achieves SOTA visual quality by eliminating artifacts caused by inaccurate masks: (a) Eliminate interference from inaccurate masks: Inaccurate masks cause over-masking, leading to unnatural regeneration of hair or hands, and mask leakage, resulting in artifacts like remnants of the old clothing. (b) Demonstration of VITON-HD In-the-Wild.

Abstract

Recent advancements in Virtual Try-On (VITON) have significantly improved image realism and garment detail preservation, driven by powerful text-to-image (T2I) diffusion models. However, existing methods often rely on user-provided masks, introducing complexity and performance degradation due to imperfect inputs, as shown in Fig.~\ref{fig:cover_image}(a). To address this, we propose a Mask-Free VITON (MF-VITON) framework that achieves realistic VITON using only a single person image and a target garment, eliminating the requirement for auxiliary masks. Our approach introduces a novel two-stage pipeline: (1) We leverage existing Mask-based VITON models to synthesize a high-quality dataset. This dataset contains diverse, realistic pairs of person images and corresponding garments, augmented with varied backgrounds to mimic real-world scenarios. (2) The pre-trained Mask-based model is fine-tuned on the generated dataset, enabling garment transfer without mask dependencies. This stage simplifies the input requirements while preserving garment texture and shape fidelity. Our framework achieves state-of-the-art (SOTA) performance regarding garment transfer accuracy and visual realism. Notably, the proposed Mask-Free model significantly outperforms existing Mask-based approaches, setting a new benchmark and demonstrating a substantial lead over previous approaches.

MF-VITON Teaser

Overview of MF-VITON: We propose a Mask-based & Mask-Free VITON pipeline that enables seamless adaptation from Mask-based to MF-VITON. The pipeline comprises two branches:

(a) the Garment Extractor, which leverages ReferenceNet to encode fine-grained garment features ℱ(xg) and employs an Adapter [Ye et al., 2023] to extract high-level semantics from garment images Xg using a pretrained image encoder;

(b) the Denoising Network, which utilizes TryonNet as the primary denoising branch to process concatenated inputs of noised latent Xt and selectively integrates either Mask-based conditions (ℱ(XMasked-Con), Mask-based text prompt) or Mask-Free conditions (ℱ(XUnmasked-Con), Mask-Free text prompt).

dataset_generation Teaser

Overview of MF-VITON Dataset Generation:

(a) In-the-Wild Mask-Free Dataset Generation: Uses FLUX.1-Fill-dev to generate realistic background-filled model images Xbg, which are then composited with the Mask-based background bg to create Mask-Free dataset samples XUnmasked-Con-bg.

(b) Mask-Free Dataset Generation: Concatenates the noised latent encoding ℱ(Xmodel) with Mask-based conditions ℱ(XMasked-Cons). The Mask-based VITON model then synthesizes garment-swapped images XUnmasked-Con.

BibTeX

@article{wan2025mfviton,
  title={MF-VITON: High-Fidelity Mask-Free Virtual Try-On with Minimal Input},
  author={Wan, Zhenchen and Xu, Yanwu and Hu, Dongting and Cheng, Weilun and Chen, Tianxi and Wang, Zhaoqing and Liu, Feng and Liu, Tongliang and Gong, Mingming},
  journal={arXiv preprint arXiv:2503.08650},
  year={2025}
}