Research map / Video / Video segmentation and tracking
Spatial-Temporal Correlation and Topology Learning for Person Re-Identification in Videos
Check 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 |
|---|---|---|---|
| add context-reinforced topology and graph convolution to the backbone and key-points estimator (ablation) | CTL · MARS | Rank-1: 88.6 -> 89.4 | Helped |
| add context-reinforced topology and graph convolution to the backbone and key-points estimator (ablation) | CTL · MARS | mAP: 82.7 -> 85.4 | Helped |
| add 3D-GCL for direct cross-spacetime information propagation (ablation) | CTL with context-reinforced topology · MARS | Rank-1: 89.4 -> 90.7 | Helped |
| add 3D-GCL for direct cross-spacetime information propagation (ablation) | CTL with context-reinforced topology · MARS | mAP: 85.4 -> 86.0 | Helped |
| add CS-GCL to fuse features across scales (ablation) | CTL with context-reinforced topology and 3D-GCL · MARS | Rank-1: 90.7 -> 91.4 | Helped |
| add CS-GCL to fuse features across scales (ablation) | CTL with context-reinforced topology and 3D-GCL · MARS | mAP: 86.0 -> 86.7 | Helped |
| add learnable mask adjacency to physical adjacency (ablation) | CTL topology · MARS | Rank-1: 90.4 -> 90.9 | Helped |
| add global-context adjacency to physical and mask adjacency (ablation) | CTL topology · MARS | Rank-1: 90.9 -> 91.4 | Helped |
| use two CS-GCL layers instead of one (ablation) | CTL · MARS | Rank-1: 91.4 -> 90.1 | Hurt |
| use two CS-GCL layers instead of one (ablation) | CTL · MARS | mAP: 86.7 -> 85.5 | Hurt |
| include s1- and s2-scale features in addition to s3-scale features (ablation) | CTL cross-scale fusion · MARS | Rank-1: 90.4 -> 91.4 | Helped |
| include s2-scale features along with s3- and s1-scale features (ablation) | CTL cross-scale fusion · MARS | mAP: 85.8 -> 86.7 | Helped |
Closest papers by meaning
- Spatial-Temporal Graph Convolutional Network for Video-Based Person Re-Identification (CVPR 2020 · similarity 0.93)
- See the Forest for the Trees: Joint Spatial and Temporal Recurrent Neural Networks for Video-Based Person Re-Identification (CVPR 2017 · similarity 0.91)
- Relation-Guided Spatial Attention and Temporal Refinement for Video-Based Person Re-Identification (AAAI 2020 · similarity 0.89)
- BiCnet-TKS: Learning Efficient Spatial-Temporal Representation for Video Person Re-Identification (CVPR 2021 · similarity 0.89)
- Spatial and Temporal Mutual Promotion for Video-Based Person Re-Identification (AAAI 2019 · similarity 0.89)
- Keypoint Message Passing for Video-Based Person Re-identification (AAAI 2022 · similarity 0.89)
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.