Research map / Large language models / Retrieval and knowledge
RAGDiffusion: Faithful Cloth Generation via External Knowledge Assimilation
Yuhan Li, Xianfeng Tan, Wenxiang Shang, Yubo Wu, Jian Wang, Xuanhong Chen, Yi Zhang, Hangcheng Zhu, Bingbing Ni
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 retrieved silhouette landmarks and structure embeddings to ReferenceNet baseline (ablation) | RAGDiffusion · STGarment test | SSIM: 0.6576 -> 0.6926 | Improved |
| add structure embeddings to EP-Adapter (ablation) | Vanilla Model · STGarment test | FID: 10.65 -> 10.55 | Improved |
| add retrieved silhouette landmarks (ablation) | Vanilla + emb · STGarment test | KID: 1.354 -> 0.940 | Improved |
| add PGEA (SDXL UNet adapted to FLUX VAE) (ablation) | + emb + sil · STGarment test | FID: 10.22 -> 9.990 | Improved |
| predict landmarks with a UNet instead of retrieving them (ablation) | RAGDiffusion · STGarment test | FID: 9.990 -> 14.55 | Hurt |
| increase memory database from 4000 to 8000 (ablation) | retrieval · STGarment retrieval | top-1 accuracy (%): 80.6 -> 79.7 | No clear effect |
| increase retrieved neighbors K from 4 to 8 (ablation) | SLLE · STGarment test | FID: 9.990 -> 10.01 | No clear effect |
| reduce K from 4 to 1 (ablation) | SLLE · STGarment test | SSIM: 0.6963 -> 0.6872 | Hurt |
| extend memory database with lower-body garments (retrieval, no retraining) | RAGDiffusion · DressCode lower-body | FID: 124.5 -> 40.5 | Improved |
| RAGDiffusion versus TryOffAnyone, evaluated on Viton-HD without training on it | RAGDiffusion · Viton-HD | FID: 12.3 vs 11.4 | No clear effect |
Closest papers by meaning
- DreamFit: Garment-Centric Human Generation via a Lightweight Anything-Dressing Encoder (AAAI 2025 · similarity 0.88)
- FashionR2R: Texture-preserving Rendered-to-Real Image Translation with Diffusion Models (NeurIPS 2024 · similarity 0.88)
- AnyDressing: Customizable Multi-Garment Virtual Dressing via Latent Diffusion Models (CVPR 2025 · similarity 0.87)
- FashionTailor: Controllable Clothing Editing for Human Images with Appearance Preserving (AAAI 2025 · similarity 0.87)
- DressWild: Feed-Forward Pose-Agnostic Garment Sewing Pattern Generation from In-the-Wild Images (NeurIPS 2026 · similarity 0.86)
- FreeCloth: Free-form Generation Enhances Challenging Clothed Human Modeling (CVPR 2025 · similarity 0.86)
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.
