VTONGuard: Automatic Detection and Authentication of AI-Generated Virtual Try-On Content

2026-01-20

AI Summary

This work addresses the growing realism of virtual try-on (VTON) generation and the associated challenges concerning authenticity and responsible usage. We present VTONGuard, the first large-scale benchmark dataset comprising 775,000 real and synthetic try-on images, spanning diverse poses, backgrounds, and clothing styles, along with a unified evaluation protocol. Building upon this foundation, we propose a multi-task detection framework that integrates auxiliary segmentation to enhance boundary-aware feature learning and cross-paradigm generalization. Extensive experiments demonstrate that our method achieves state-of-the-art overall performance on VTONGuard, while systematically revealing the strengths, limitations, and generalization bottlenecks of existing detection approaches. This study provides both the data infrastructure and methodological insights necessary for advancing trustworthy VTON technologies.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

With the rapid advancement of generative AI, virtual try-on (VTON) systems are becoming increasingly common in e-commerce and digital entertainment. However, the growing realism of AI-generated try-on content raises pressing concerns about authenticity and responsible use. To address this, we present VTONGuard, a large-scale benchmark dataset containing over 775,000 real and synthetic try-on images. The dataset covers diverse real-world conditions, including variations in pose, background, and garment styles, and provides both authentic and manipulated examples. Based on this benchmark, we conduct a systematic evaluation of multiple detection paradigms under unified training and testing protocols. Our results reveal each method's strengths and weaknesses and highlight the persistent challenge of cross-paradigm generalization. To further advance detection, we design a multi-task framework that integrates auxiliary segmentation to enhance boundary-aware feature learning, achieving the best overall performance on VTONGuard. We expect this benchmark to enable fair comparisons, facilitate the development of more robust detection models, and promote the safe and responsible deployment of VTON technologies in practice.

Problem

• virtual try-on
• AI-generated content
• authenticity
• content detection
• generative AI

Innovation

• virtual try-on
• AI-generated content detection
• benchmark dataset
• multi-task learning
• boundary-aware feature learning