State of the Field
Recent advancements in virtual try-on technology are focusing on enhancing the realism and efficiency of garment fitting solutions for online retail, addressing key commercial challenges such as high return rates and customer dissatisfaction. New frameworks, like BridgeDiff, are bridging the gap between on-body appearances and flat-garment representations, improving the accuracy of garment synthesis through innovative modules that enhance structural stability and detail preservation. Additionally, the introduction of error enumeration techniques in reinforcement learning is refining evaluation metrics, allowing for more nuanced assessments of garment fit and appearance, which is crucial in a domain where subtle errors can significantly impact consumer perception. Datasets like MV-Fashion are providing rich, multi-view data to train models on complex garment dynamics, while culturally diverse datasets are expanding the applicability of virtual try-on systems beyond Western clothing. Collectively, these efforts are pushing the field toward more robust, user-friendly solutions that cater to a broader range of consumer needs.
Papers
1–6 of 6BridgeDiff: Bridging Human Observations and Flat-Garment Synthesis for Virtual Try-Off
Virtual try-off (VTOFF) aims to recover canonical flat-garment representations from images of dressed persons for standardized display and downstream virtual try-on. Prior methods often treat VTOFF as...
When Rubrics Fail: Error Enumeration as Reward in Reference-Free RL Post-Training for Virtual Try-On
Reinforcement learning with verifiable rewards (RLVR) and Rubrics as Rewards (RaR) have driven strong gains in domains with clear correctness signals and even in subjective domains by synthesizing eva...
MV-Fashion: Towards Enabling Virtual Try-On and Size Estimation with Multi-View Paired Data
Existing 4D human datasets fall short for fashion-specific research, lacking either realistic garment dynamics or task-specific annotations. Synthetic datasets suffer from a realism gap, whereas real-...
Virtual Try-On for Cultural Clothing: A Benchmarking Study
Although existing virtual try-on systems have made significant progress with the advent of diffusion models, the current benchmarks of these models are based on datasets that are dominant in western-s...
PROMO: Promptable Outfitting for Efficient High-Fidelity Virtual Try-On
Virtual Try-on (VTON) has become a core capability for online retail, where realistic try-on results provide reliable fit guidance, reduce returns, and benefit both consumers and merchants. Diffusion-...
VTEdit-Bench: A Comprehensive Benchmark for Multi-Reference Image Editing Models in Virtual Try-On
As virtual try-on (VTON) continues to advance, a growing number of real-world scenarios have emerged, pushing beyond the ability of the existing specialized VTON models. Meanwhile, universal multi-ref...