Research map / Detection and segmentation / CNN features
Human Semantic Parsing for Person Re-Identification
Read the paperCheck 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 |
|---|---|---|---|
| use semantic body-region pooling and combine it with the global representation | Inception-V3 person re-identification · Market-1501 | mAP(%): 73.06 -> 79.67 | Helped |
| use semantic body-region pooling and combine it with the global representation | Inception-V3 person re-identification · CUHK03 | rank-1: 87.91 -> 91.21 | Helped |
| use semantic body-region pooling and combine it with the global representation | Inception-V3 person re-identification · DukeMTMC-reID | mAP(%): 59.87 -> 68.78 | Helped |
| exclude the foreground representation (ablation) | SPReID · Market-1501 | mAP(%): 79.67 -> 78.06 | Hurt |
| exclude the foreground representation (ablation) | SPReID · CUHK03 | rank-1: 91.21 -> 91.29 | No clear effect |
| fine-tune using higher-resolution 748×246 input images (ablation) | Model-L · Market-1501 | mAP(%): 73.06 -> 76.56 | Helped |
| fine-tune using higher-resolution 748×246 input images (ablation) | Model-L · CUHK03 | rank-1: 87.91 -> 88.73 | Helped |
| fine-tune using higher-resolution 748×246 input images (ablation) | Model-L · DukeMTMC-reID | mAP(%): 59.87 -> 63.27 | Helped |
| increase input resolution from 246×82 to 375×125 (ablation) | Inception-V3 baseline · Market-1501 | mAP(%): 64.45 -> 72.14 | Helped |
| replace Inception-V3 with ResNet-152 (ablation) | baseline re-identification model · DukeMTMC-reID | mAP(%): 59.87 -> 62.42 | Helped |
| share the backbone between the two aggregation heads (ablation) | SPReID w/fg · Market-1501 | mAP(%): 78.66 -> 77.62 | Hurt |
| share the backbone between the two aggregation heads (ablation) | SPReID w/fg · CUHK03 | rank-1: 89.57 -> 87.69 | Hurt |
Closest papers by meaning
- Semantics-Aligned Representation Learning for Person Re-Identification (AAAI 2020 · similarity 0.90)
- Resource Aware Person Re-Identification Across Multiple Resolutions (CVPR 2018 · similarity 0.90)
- Densely Semantically Aligned Person Re-Identification (CVPR 2019 · similarity 0.90)
- Deep Neural Networks with Inexact Matching for Person Re-Identification (NeurIPS 2016 · similarity 0.88)
- Dynamic Feature Pruning and Consolidation for Occluded Person Re-identification (AAAI 2024 · similarity 0.88)
- FlexiReID: Adaptive Mixture of Expert for Multi-Modal Person Re-Identification (ICML 2025 · similarity 0.88)
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.