Research map / Video / Video segmentation and tracking
Learning Instance-Level Spatial-Temporal Patterns for Person Re-Identification
Min Ren, Lingxiao He, Xingyu Liao, Wu Liu, Yunlong Wang, Tieniu Tan
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 instance-level walking-direction state information (ablation) | InSTD · DukeMTMC-reID | mAP: 87.1% -> 89.1% | Improved |
| add instance-level walking-direction state information (ablation) | InSTD · DukeMTMC-reID | Rank-1: 94.3% -> 95.7% | Improved |
| add decoupling of spatial and temporal patterns (ablation) | InSTD · DukeMTMC-reID | mAP: 86.9% -> 89.1% | Improved |
| add decoupling of spatial and temporal patterns (ablation) | InSTD · DukeMTMC-reID | Rank-1: 95.0% -> 95.7% | Improved |
| add both instance info and ST decoupling (ablation) | InSTD · DukeMTMC-reID | mAP: 83.4% -> 89.1% | Improved |
| add both instance info and ST decoupling (ablation) | InSTD · DukeMTMC-reID | Rank-1: 93.8% -> 95.7% | Improved |
| add InSTD spatial-temporal constraints to ResNet-50 baseline | ResNet-50 · Market-1501 | mAP: 82.2% -> 90.8% | Improved |
| add InSTD spatial-temporal constraints to ResNet-50 baseline | ResNet-50 · DukeMTMC-reID | mAP: 72.7% -> 89.1% | Improved |
| add InSTD to PCB feature extractor | PCB · DukeMTMC-reID | mAP: 66.1% -> 86.1% | Improved |
| add InSTD to VPM feature extractor | VPM · DukeMTMC-reID | mAP: 72.6% -> 89.3% | Improved |
| replace st-ReID with InSTD | ResNet-50 · Market-1501 | mAP: 88.6% -> 90.8% | Improved |
| replace st-ReID with InSTD | ResNet-50 · DukeMTMC-reID | mAP: 84.3% -> 89.1% | Improved |
Closest papers by meaning
- Spatial-Temporal Person Re-Identification (AAAI 2019 · similarity 0.91)
- A Systematic Evaluation and Benchmark for Person Re-Identification: Features, Metrics, and Datasets (TPAMI 2018 · similarity 0.90)
- Spatio-Temporal Representation Factorization for Video-Based Person Re-Identification (ICCV 2021 · similarity 0.90)
- Person Re-Identification in the Wild (CVPR 2017 · similarity 0.90)
- Spatial and Temporal Mutual Promotion for Video-Based Person Re-Identification (AAAI 2019 · similarity 0.90)
- From Poses to Identity: Training-Free Person Re-Identification via Feature Centralization (CVPR 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.
