Research map / Label-efficient and robust learning / Recognition and long-tail
DeepFashion: Powering Robust Clothes Recognition and Retrieval With Rich Annotations
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| jointly predict clothing landmarks and attributes and pool or gate features at the landmarks | FashionNet · DeepFashion category classification | category top-5 accuracy: 82.58 -> 90.17 | Helped |
| use FashionNet with landmark pooling and attribute supervision | DeepFashion in-shop clothes retrieval · in-shop clothes retrieval | top-20 retrieval accuracy: 0.675 to 0.764 | Helped |
| use FashionNet with landmark pooling and attribute supervision | DeepFashion consumer-to-shop retrieval · consumer-to-shop clothes retrieval | top-20 retrieval accuracy: 0.188 | Helped |
| add clothing-specific landmarks instead of using human joints (ablation) | FashionNet · DeepFashion category classification | category accuracy: 6∼9 percent performance drops | Helped |
| add clothing-specific landmarks instead of using poselets (ablation) | FashionNet · DeepFashion in-shop clothes retrieval | retrieval accuracy: 8 and 6 percent | Helped |
| use 1,000 attributes instead of 500 (ablation) | FashionNet · DeepFashion category classification | category top-5 accuracy: 13 percent | Helped |
| use 1,000 attributes instead of 100 (ablation) | FashionNet · DeepFashion category classification | category top-5 accuracy: 20 percent | Helped |
| add more training attributes rather than use fewer attributes (ablation) | FashionNet · DeepFashion in-shop clothes retrieval | retrieval accuracy: 19 and 12 percent | Helped |
| add more training attributes rather than use fewer attributes (ablation) | FashionNet · DeepFashion consumer-to-shop clothes retrieval | retrieval accuracy | Helped |
| use clothing landmarks instead of human landmarks (ablation) | FashionNet · DeepFashion consumer-to-shop clothes retrieval | retrieval accuracy | Helped |
| use attribute information to assist landmark localization (ablation) | FashionNet · DeepFashion landmark estimation | landmark detection rate | Helped |
Closest papers by meaning
- AnyDressing: Customizable Multi-Garment Virtual Dressing via Latent Diffusion Models (CVPR 2025 · similarity 0.85)
- AIpparel: A Multimodal Foundation Model for Digital Garments (CVPR 2025 · similarity 0.85)
- FashionTailor: Controllable Clothing Editing for Human Images with Appearance Preserving (AAAI 2025 · similarity 0.84)
- FashionERN: Enhance-and-Refine Network for Composed Fashion Image Retrieval (AAAI 2024 · similarity 0.84)
- ChatGarment: Garment Estimation, Generation and Editing via Large Language Models (CVPR 2025 · similarity 0.84)
- MV-VTON: Multi-View Virtual Try-On with Diffusion Models (AAAI 2025 · similarity 0.84)
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.