Research map / Image generation / Editing and style transfer
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
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| give detailed text (shape/pose plus layering/styling) in addition to item category (ablation) | Nano Banana · Garments2Look test (sampled subset) | FID: 23.272 -> 21.545 | Improved |
| add detailed text annotations to item category (ablation) | Nano Banana · Garments2Look test (sampled subset) | KID: 1.271 -> 0.641 | Improved |
| add outfit overall description to item category text (ablation) | Nano Banana · Garments2Look test (sampled subset) | SSIM: 0.817 -> 0.810 | Hurt |
| feed outfit as one OOTD grid image (2 Ref) instead of N separate garment images (ablation) | Nano Banana · Garments2Look test | Garment consistency: 0.8045 -> 0.9253 | Improved |
| OOTD grid input instead of N separate images (ablation) | Nano Banana · Garments2Look test | Layering accuracy: 0.9250 -> 0.8847 | Hurt |
| OOTD grid input instead of N separate images (ablation) | Qwen-Image-Edit-2509 · Garments2Look test | Garment consistency: 0.5144 -> 0.7940 | Improved |
| OOTD grid input instead of N separate images (ablation) | Nano Banana Pro · Garments2Look test | Garment consistency: 0.9836 -> 0.9707 | No clear effect |
| use general image editing model (Nano Banana) instead of VTON-specific FastFit on Garments2Look | outfit-level VTON · Garments2Look test | Layering accuracy: 0.1305 -> 0.8847 | Improved |
| use Nano Banana instead of FastFit on the older DressCode-MR benchmark | multi-reference VTON · DressCode-MR (paired) | FID: 9.311 -> 15.220 | Hurt |
| rearrange item positions in the OOTD grid (ablation) | Nano Banana look synthesis · Garments2Look synthesis | look quality | No clear effect |
Closest papers by meaning
- MV-VTON: Multi-View Virtual Try-On with Diffusion Models (AAAI 2025 · similarity 0.91)
- M&M VTO: Multi-Garment Virtual Try-On and Editing (CVPR 2024 · similarity 0.91)
- AnyFit: Controllable Virtual Try-on for Any Combination of Attire Across Any Scenario (NeurIPS 2024 · similarity 0.91)
- IMAGDressing-v1: Customizable Virtual Dressing (AAAI 2025 · similarity 0.91)
- OpenVTON-Bench: A Large-Scale High-Resolution Benchmark for Controllable Virtual Try-On Evaluation (NeurIPS 2026 · similarity 0.90)
- VTON 360: High-Fidelity Virtual Try-On from Any Viewing Direction (CVPR 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
