Research map / Image generation / Editing and style transfer
Towards Scalable Unpaired Virtual Try-On via Patch-Routed Spatially-Adaptive GAN
Zhenyu Xie, Zaiyu Huang, Fuwei Zhao, Haoye Dong, Michael Kampffmeyer, Xiaodan Liang
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 |
|---|---|---|---|
| Use PASTA-GAN instead of the strongest listed baseline, Liquid Warping GAN | Unpaired virtual try-on · DeepFashion | FID: 42.18 -> 21.58 | Improved |
| Use PASTA-GAN instead of the strongest listed baseline, ADGAN | Unpaired virtual try-on · DeepFashion | Human Evaluation: 21.29% -> 54.27% | Improved |
| Use PASTA-GAN instead of the strongest listed baseline, Liquid Warping GAN | Unpaired virtual try-on · UPT | FID: 33.18 -> 7.852 | Improved |
| Use PASTA-GAN instead of the strongest listed baseline, Liquid Warping GAN | Unpaired virtual try-on · UPT | Human Evaluation: 9.310% -> 74.22% | Improved |
| Use PASTA-GAN instead of PFAFN | Virtual try-on in paired benchmark setting · MPV | FID: 17.40 -> 16.48 | Improved |
| Use PASTA-GAN instead of PFAFN | Virtual try-on in paired benchmark setting · MPV | Human Evaluation: 28.71% -> 50.57% | Improved |
| Replace intact garment input with normalized garment patches in the style encoder (ablation) | PASTA-GAN style synthesis without the texture synthesis branch · Try-on results (qualitative ablation) | Garment shape accuracy | Improved |
| Add the texture synthesis branch and warped-garment texture features (ablation) | PASTA-GAN · Try-on results (qualitative ablation) | Garment texture detail | Improved |
| Add the style synthesis branch (ablation) | PASTA-GAN texture synthesis configuration · Try-on results (qualitative ablation) | Garment alignment | Improved |
| Replace normal residual blocks with spatially-adaptive residual blocks (ablation) | PASTA-GAN texture synthesis branch · Try-on results (qualitative ablation) | Garment texture quality | Improved |
| Apply random erasing to the warped garment during training (ablation) | PASTA-GAN · Unpaired virtual try-on (random-erasing ablation) | Try-on quality | No clear effect |
Closest papers by meaning
- FW-GAN: Flow-Navigated Warping GAN for Video Virtual Try-On (ICCV 2019 · similarity 0.92)
- Towards Multi-Pose Guided Virtual Try-On Network (ICCV 2019 · similarity 0.92)
- Toward Characteristic-Preserving Image-based Virtual Try-On Network (ECCV 2018 · similarity 0.92)
- MV-VTON: Multi-View Virtual Try-On with Diffusion Models (AAAI 2025 · similarity 0.90)
- CycleVTON: A Cycle Mapping Framework for Parser-Free Virtual Try-On (AAAI 2024 · similarity 0.90)
- Enhancing Virtual Try-On with Synthetic Pairs and Error-Aware Noise Scheduling (CVPR 2025 · similarity 0.89)
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.
