Research map / Image generation / Editing and style transfer
Dressing in Order: Recurrent Person Image Generation for Pose Transfer, Virtual Try-On and Outfit Editing
Aiyu Cui, Daniel McKee, Svetlana Lazebnik
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 |
|---|---|---|---|
| add recurrent garment-by-garment generation (ablation) | DiOr small · DeepFashion pose transfer (256x176, ablation of small model) | sIoU: 58.44 -> 58.99 | Improved |
| add recurrent generation (ablation) | DiOr small · DeepFashion pose transfer (256x176, ablation of small model) | FID: 14.85 -> 14.34 | Improved |
| train with inpainting instead of pose transfer only (ablation) | DiOr small · DeepFashion pose transfer (256x176, ablation of small model) | FID: 13.77 -> 14.34 | No clear effect |
| train with inpainting instead of reconstruction (ablation) | DiOr small · DeepFashion pose transfer (256x176, ablation of small model) | sIoU: 58.33 -> 58.99 | No clear effect |
| use separate 2D shape and texture encoding instead of single 2D encoding (ablation) | DiOr small · DeepFashion pose transfer (256x176, ablation of small model) | FID: 15.18 -> 14.34 | Improved |
| use 2D instead of 1D single encoding (ablation) | DiOr small · DeepFashion pose transfer (256x176, ablation of small model) | FID: 16.14 -> 14.34 | Improved |
| use 2D separate shape and texture instead of 1D texture+1D shape (ablation) | DiOr small · DeepFashion pose transfer (256x176, ablation of small model) | FID: 20.07 -> 14.34 | Improved |
| use 2D shape rather than 1D shape with 1D texture (ablation) | DiOr small · DeepFashion pose transfer (256x176, ablation of small model) | sIoU: 57.40 -> 58.99 | Improved |
| add flow field estimator (ablation) | DiOr small · DeepFashion pose transfer (256x176, ablation of small model) | FID: 16.47 -> 14.34 | Improved |
| add flow field estimator (ablation) | DiOr small · DeepFashion pose transfer (256x176, ablation of small model) | sIoU: 56.28 -> 58.99 | Improved |
| use large model (L=256) instead of small (L=128) | DiOr · DeepFashion pose transfer (256x176) | FID: 14.34 -> 13.59 | Improved |
| use large model instead of small | DiOr · DeepFashion pose transfer (256x256) | sIoU: 57.22 -> 58.63 | Improved |
Closest papers by meaning
- FashionChameleon: Towards Real-Time and Interactive Human-Garment Video Customization (NeurIPS 2026 · similarity 0.88)
- Correpondence Alignment For Improved Virtual Try-On (NeurIPS 2026 · similarity 0.87)
- IMAGDressing-v1: Customizable Virtual Dressing (AAAI 2025 · similarity 0.87)
- FashionTailor: Controllable Clothing Editing for Human Images with Appearance Preserving (AAAI 2025 · similarity 0.87)
- AnyDressing: Customizable Multi-Garment Virtual Dressing via Latent Diffusion Models (CVPR 2025 · similarity 0.87)
- Controllable Human Image Generation with Personalized Multi-Garments (CVPR 2025 · similarity 0.87)
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.
