DS-VTON: High-Quality Virtual Try-on via Disentangled Dual-Scale Generation

2025-06-01

AI Summary

Existing virtual try-on methods struggle to simultaneously achieve precise garment-human geometric alignment and high-fidelity texture reconstruction. To address this, we propose a decoupled dual-scale generative framework: first modeling cross-domain semantic correspondences at low resolution, then reconstructing fine-grained textures and structural details at high resolution via a residual-guided diffusion model. Our approach is the first fully mask-free, end-to-end virtual try-on solution—eliminating reliance on human parsing maps and instead leveraging built-in semantic priors from pre-trained diffusion models to ensure appearance consistency and pose robustness. Evaluated on multiple standard benchmarks, our method achieves state-of-the-art performance in both structural alignment and texture fidelity, significantly improving image naturalness, detail sharpness, and pose consistency.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Despite recent progress, most existing virtual try-on methods still struggle to simultaneously address two core challenges: accurately aligning the garment image with the target human body, and preserving fine-grained garment textures and patterns. In this paper, we propose DS-VTON, a dual-scale virtual try-on framework that explicitly disentangles these objectives for more effective modeling. DS-VTON consists of two stages: the first stage generates a low-resolution try-on result to capture the semantic correspondence between garment and body, where reduced detail facilitates robust structural alignment. The second stage introduces a residual-guided diffusion process that reconstructs high-resolution outputs by refining the residual between the two scales, focusing on texture fidelity. In addition, our method adopts a fully mask-free generation paradigm, eliminating reliance on human parsing maps or segmentation masks. By leveraging the semantic priors embedded in pretrained diffusion models, this design more effectively preserves the person's appearance and geometric consistency. Extensive experiments demonstrate that DS-VTON achieves state-of-the-art performance in both structural alignment and texture preservation across multiple standard virtual try-on benchmarks.

Problem

• Accurately aligning garment with target human body
• Preserving fine-grained garment textures and patterns
• Eliminating reliance on human parsing or segmentation masks

Innovation

• Dual-scale framework for alignment and texture
• Residual-guided diffusion for high-resolution refinement
• Mask-free generation using semantic priors