Virtual Try-On for Cultural Clothing: A Benchmarking Study

Muhammad Tausif Ul Islam; Shahir Awlad; Sameen Yeaser Adib; Md. Atiqur Rahman; Sabbir Ahmed; Md. Hasanul Kabir

cs.CVMar 7, 2026

Virtual Try-On for Cultural Clothing: A Benchmarking Study

Authors: Muhammad Tausif Ul Islam, Shahir Awlad, Sameen Yeaser Adib, Md. Atiqur Rahman, Sabbir Ahmed, Md. Hasanul Kabir

Abstract

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-style clothing and female models, limiting their ability to generalize culturally diverse clothing styles. In this work, we introduce BD-VITON, a virtual try-on dataset focused on Bangladeshi garments, including saree, panjabi and salwar kameez, covering both male and female categories as well. These garments present unique structural challenges such as complex draping, asymmetric layering, and high deformation complexities which are underrepresented in the original VITON dataset. To establish strong baselines, we retrain and evaluate try-on models, namely StableViton, HR-VITON, and VITON-HD on our dataset. Our experiments demonstrate consistent improvements in terms of both quantitative and qualitative analysis, compared to zero shot inference.

Figures & tables

Panel of Figure 1 — Fig. 1 : Samples of BD-VITON, Kameez, Panjabi and Saree from left to right

Clothing Type	Male Samples	Female Samples
Panjabi	340	-
Kameez	-	341
Saree	-	332
Total	340	673

TABLE I : BD-VITON dataset composition by gender and attire type.

Fig. 2 : Resolution distribution of BD-VITON (Train+Test).

Dataset	Pairs	Median Res.	Gender	Body Coverage	Garment Types	Cultural Scope
VITON	16,253	256 $\times$ 192	F only	Upper-body	Tops, shirts	Western
VITON-HD	13,679	1024 $\times$ 768	F only	Upper-body	Tops, shirts	Western
DressCode	53,792	1024 $\times$ 768	M+F	Full-body	Tops, bottoms, t shirts, skirts	Western
BD-VITON (Ours)	1,013	1080 $\times$ 1440	M+F	Full-body	Saree, Panjabi, Kameez	Bangladeshi

TABLE II : Comparison of BD-VITON with existing virtual try-on datasets.

Panel of Figure 5 — Fig. 3 : BD-VITON composition example. Top row: Original Image $I_{org}$ , Parse $I_{parse}$ , Parse-Agnostic $I_{parseAgn}$ , Warped Cloth Mask $I_{wClothM}$ , Cloth $I_{cloth}$ . Bottom row: Cloth Mask $I_{ClothM}$ , Agnostic Mask $I_{AgnM}$ , Agnostic $I_{Agn}$ , Densepose $I_{dp}$ , OpenPose Rendering $I_{op}$ .

Pose Attribute	Value	Remarks
Frontal Pose	87.16%	Subject facing camera
Slightly Turned Pose	12.84%	Minor body rotation
Avg. Visible Keypoints	17.62/18	Confidence $\geq$ 0.2

TABLE III : Pose Diversity of BD-VITON

Model	Zero-Shot	SSIM $\uparrow$	LPIPS $\downarrow$	FID $\downarrow$
HR-VITON	✓	0.478	0.675	259.25
HR-VITON	✗	0.815	0.156	42.92
StableVITON	✓	0.722	0.238	69.59
StableVITON	✗	0.732	0.219	50.40
VITON-HD	✓	0.813	0.244	100.07
VITON-HD	✗	0.868	0.154	49.89

TABLE IV: Quantitative Results on BD-VITON’s test set (203 images).

Panel of Figure 8 — Fig. 4 : Qualitative comparison of virtual try-on results across three garment types: saree (row 1), panjabi (row 2), and kameez (row 3). Each method shows zero-shot (left) and trained (right) results.

Virtual Try-On

Virtual try-on generates images or videos showing people wearing selected garments while preserving appearance, fit, pose, and identity.

See all 55 papers

Garments2Look: A Multi-Reference Dataset for High-Fidelity Outfit-Level Virtual Try-On with Clothing and Accessories
Mar 14, 2026Junyao Hu, Zhongwei Cheng, Waikeung Wong +1Virtual Try-On
The Hong Kong Polytechnic University · Huhu AI Inc.
FitVTON: Fit-aware Virtual Try-On via Body-Garment Size Control
Jun 10, 2026Yiqun Ning, Ao Shen, Chenhang He +1Virtual Try-On
Department of Computing, The Hong Kong Polytechnic University, Hong Kong · Nuvatech
Beyond Global Realism: Virtual Try-On Evaluation and Optimization with Dimension-wise Garment Fidelity Assessment
Aug 30, 2026Kaidong Zhang, Yukang Ding, Xiaoyu Liu +1Virtual Try-On
Taobao & Tmall Group of Alibaba · Harbin Institute of Technology

Mar 14, 2026J/K move · Enter open · S save

Garments2Look: A Multi-Reference Dataset for High-Fidelity Outfit-Level Virtual Try-On with Clothing and Accessories

Junyao Hu, Zhongwei Cheng, Waikeung Wong, Xingxing Zou

The Hong Kong Polytechnic University · Huhu AI Inc.

Virtual Try-On

Open paper AskPDF arXiv

Virtual try-on (VTON) has advanced single-garment visualization, yet real-world fashion centers on full outfits with multiple garments, accessories, fine-grained categories, layering, and diverse styling, remaining beyond current VTON systems. Existing datasets are category-limited and lack outfit diversity. We introduce Garments2Look, the first large-scale multimodal dataset for outfit-level VTON, comprising 80K many-garments-to-one-look pairs across 40 major categories and 300+ fine-grained subcategories. Each pair includes an outfit with 3-12 reference garment images (Average 4.48), a model image wearing the outfit, and detailed item and try-on textual annotations. To balance authenticity and diversity, we propose a synthesis pipeline. It involves heuristically constructing outfit lists before generating try-on results, with the entire process subjected to strict automated filtering and human validation to ensure data quality. To probe task difficulty, we adapt SOTA VTON methods and general-purpose image editing models to establish baselines. Results show current methods struggle to try on complete outfits seamlessly and to infer correct layering and styling, leading to misalignment and artifacts.

Abstract

Virtual Try-On

Explore similar work

Garments2Look: A Multi-Reference Dataset for High-Fidelity Outfit-Level Virtual Try-On with Clothing and Accessories

FitVTON: Fit-aware Virtual Try-On via Body-Garment Size Control

Beyond Global Realism: Virtual Try-On Evaluation and Optimization with Dimension-wise Garment Fidelity Assessment