Research map / Vision-language and multimodal / Human and embodied modeling
Automatic Spatially-Aware Fashion Concept Discovery
Xintong Han, Zuxuan Wu, Phoenix X. Huang, Xiao Zhang, Menglong Zhu, Yuan Li, Yang Zhao, Larry S. Davis
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 spatial representation to semantic word vectors when clustering attributes (ablation) | concept discovery · Fashion200K attribute clustering | V-measure: 0.629 -> 0.658 | Improved |
| add semantic word vectors to spatial-only attribute representation (ablation) | concept discovery · Fashion200K attribute clustering | V-measure: 0.540 -> 0.658 | Improved |
| joint semantic+spatial clustering instead of ACD (semantic plus visual similarity) | concept discovery · Fashion200K attribute clustering | V-measure: 0.626 -> 0.658 | Improved |
| add concept discovery (joint semantic+spatial negative attributes) to VSE baseline | attribute-feedback product retrieval · Fashion200K retrieval (dress, top, jacket, pants, skirt) | top-k retrieval accuracy: highest retrieval accuracy for all five categories | Improved |
| use Word2vec semantic-only concepts instead of no concept discovery (ablation) | attribute-feedback retrieval · Fashion200K retrieval (tops) | top-k retrieval accuracy | Hurt |
| add spatial information to semantic concepts for pants (ablation) | attribute-feedback retrieval · Fashion200K retrieval (pants) | top-k retrieval accuracy | No clear effect |
Closest papers by meaning
- Fashion Microscope: Pixel-Level Attribute Perception via Optimal Transport and Neural Semantic Aggregation (AAAI 2026 · similarity 0.88)
- Learning Attribute-Driven Disentangled Representations for Interactive Fashion Retrieval (ICCV 2021 · similarity 0.86)
- Learning to Appreciate the Aesthetic Effects of Clothing (AAAI 2016 · similarity 0.86)
- Fashion Retrieval via Graph Reasoning Networks on a Similarity Pyramid (ICCV 2019 · similarity 0.86)
- Easy Regional Contrastive Learning of Expressive Fashion Representations (NeurIPS 2024 · similarity 0.83)
- TIGER-FG: Text-Guided Implicit Fine-Grained Grounding for E-commerce Retrieval (NeurIPS 2026 · similarity 0.83)
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.
