Research map / Detection and segmentation / Weakly supervised detection and segmentation
Improving Pedestrian Attribute Recognition With Weakly-Supervised Multi-Scale Attribute-Specific Localization
Chufeng Tang, Lu Sheng, Zhaoxiang Zhang, Xiaolin Hu
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 ALM at a single level (incep_5b) (ablation) | BN-Inception baseline · RAP | mA: 75.76 -> 77.45 | Improved |
| add ALMs at three feature levels instead of one (ablation) | ALM single level · RAP | mA: 77.45 -> 78.89 | Improved |
| combine levels by element-wise addition (FPN-like) (ablation) | multi-level ALM · RAP | mA: 78.89 -> 78.51 | Hurt |
| combine levels by concatenation instead of addition (ablation) | multi-level ALM · RAP | mA: 78.51 -> 79.93 | Improved |
| add channel-attention sub-network (ablation) | top-down concatenation · RAP | mA: 79.93 -> 80.61 | Improved |
| use element-wise maximum instead of averaging in deep supervision voting (ablation) | ALM with deep supervision · RAP | mA: 80.70 -> 81.87 | Improved |
| add ALMs to the full model (remove all ALMs) (ablation) | full model · RAP | mA: 78.91 -> 81.87 | Improved |
| attribute-specific box localization instead of rigid body parts (ablation) | ALM framework · RAP | mA: 76.56 -> 81.87 | Improved |
| attribute-specific box localization instead of attention masks (ablation) | ALM framework · RAP | mA: 78.35 -> 81.87 | Improved |
| add the full proposed method to the BN-Inception baseline (ablation) | BN-Inception · PA-100K | mA: 77.47 -> 80.68 | Improved |
| add the full proposed method to the BN-Inception baseline | BN-Inception · PETA | mA: 82.66 -> 86.30 | Improved |
| add the full proposed method to the BN-Inception baseline | BN-Inception · RAP | mA: 75.76 -> 81.87 | Improved |
Closest papers by meaning
- Learning Disentangled Attribute Representations for Robust Pedestrian Attribute Recognition (AAAI 2022 · similarity 0.90)
- Pedestrian Attribute Recognition: A New Benchmark Dataset and a Large Language Model Augmented Framework (AAAI 2025 · similarity 0.89)
- Selective and Orthogonal Feature Activation for Pedestrian Attribute Recognition (AAAI 2024 · similarity 0.89)
- Pedestrian Attribute Recognition as Label-balanced Multi-label Learning (ICML 2024 · similarity 0.85)
- Enhanced Visual-Semantic Interaction with Tailored Prompts for Pedestrian Attribute Recognition (CVPR 2025 · similarity 0.84)
- Attribute-Guided Pedestrian Retrieval: Bridging Person Re-ID with Internal Attribute Variability (CVPR 2024 · 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.
