TripVVT: A Large-Scale Triplet Dataset and a Coarse-Mask Baseline for In-the-Wild Video Virtual Try-On

TL;DR

This paper introduces TripVVT, a large-scale in-the-wild triplet dataset and a diffusion transformer-based framework for improved, realistic, and stable video virtual try-on, addressing current limitations in data scarcity and mask usage.

Contribution

It provides the largest diverse triplet dataset, a novel human-mask prior-based framework, and a comprehensive benchmark for evaluating video virtual try-on methods.

Findings

TripVVT outperforms existing systems in video quality and garment fidelity.

The dataset and benchmark facilitate progress in realistic video virtual try-on.

The proposed method generalizes well to challenging in-the-wild scenarios.

Abstract

Due to the scarcity of large-scale in-the-wild triplet data and the improper use of masks, the performance of video virtual try-on models remains limited. In this paper, we first introduce **TripVVT-10K**, the largest and most diverse in-the-wild triplet dataset to date, providing explicit video-level cross-garment supervision that existing video datasets lack. Built upon this resource, we develop **TripVVT**, a Diffusion Transformer-based framework that replaces fragile garment masks with a simple, stable human-mask prior, enabling reliable background preservation while remaining robust to real-world motion, occlusion, and cluttered scenes. To support comprehensive evaluation, we further establish **TripVVT-Bench**, a 100-case benchmark covering diverse garments, complex environments, and multi-person scenarios, with metrics spanning video quality, try-on fidelity, background consistency, and temporal coherence. Compared to state-of-the-art academic and commercial systems, TripVVT achieves superior video quality and garment fidelity while markedly improving generalization to challenging in-the-wild videos. We publicly release the dataset and benchmark, which we believe provide a solid foundation for advancing controllable, realistic, and temporally stable video virtual try-on.