Research map / Detection and segmentation / Open-set and few-shot detection
Fashionpedia: Ontology, Segmentation, and an Attribute Localization Dataset
Menglin Jia, Mengyun Shi, Mikhail Sirotenko, Yin Cui, Claire Cardie, Bharath Hariharan, Hartwig Adam, Serge Belongie
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 macro-F1 attribute constraint to true-positive definition | Attribute-Mask R-CNN SpineNet-143 · Fashionpedia val | box AP: 48.7 -> 35.7 | Hurt |
| add macro-F1 attribute constraint | Attribute-Mask R-CNN SpineNet-143 outerwear · Fashionpedia val (outerwear) | mask AP: 64.1 -> 40.7 | Hurt |
| add macro-F1 attribute constraint | Attribute-Mask R-CNN SpineNet-143 parts · Fashionpedia val (parts) | mask AP: 19.3 -> 13.4 | Hurt |
| lengthen schedule from 1x to 3x | Attribute-Mask R-CNN R50-FPN · Fashionpedia val | mask AP (IoU): 34.3 -> 39.2 | Improved |
| lengthen schedule from 3x to 6x | Attribute-Mask R-CNN R50-FPN · Fashionpedia val | box AP (IoU): 43.4 -> 42.9 | Hurt |
| lengthen schedule from 3x to 6x | Attribute-Mask R-CNN R101-FPN · Fashionpedia val | mask AP (IoU): 40.7 -> 39.7 | Hurt |
| lengthen schedule from 3x to 6x | Attribute-Mask R-CNN R50-FPN · Fashionpedia val | mask AP (IoU+F1): 29.5 -> 30.2 | Improved |
| use R101-FPN instead of R50-FPN backbone (1x) | Attribute-Mask R-CNN · Fashionpedia val | mask AP (IoU): 34.3 -> 36.7 | Improved |
| use SpineNet-143 instead of R101-FPN 6x | Attribute-Mask R-CNN · Fashionpedia val | mask AP (IoU+F1): 31.3 -> 33.3 | Improved |
| scale SpineNet from 49 to 143 | Attribute-Mask R-CNN · Fashionpedia val | box AP (IoU): 43.7 -> 48.7 | Improved |
Closest papers by meaning
- "Fashionformer: A Simple, Effective and Unified Baseline for Human Fashion Segmentation and Recognition" (ECCV 2022 · similarity 0.89)
- Fashion Microscope: Pixel-Level Attribute Perception via Optimal Transport and Neural Semantic Aggregation (AAAI 2026 · similarity 0.88)
- Learning Attribute Representations With Localization for Flexible Fashion Search (CVPR 2018 · similarity 0.88)
- Fine-Grained Fashion Representation Learning by Online Deep Clustering (ECCV 2022 · similarity 0.88)
- Learning Attribute and Class-Specific Representation Duet for Fine-Grained Fashion Analysis (CVPR 2023 · similarity 0.88)
- Easy Regional Contrastive Learning of Expressive Fashion Representations (NeurIPS 2024 · 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.
