Research map / Video / Video segmentation and tracking
Keypoint Message Passing for Video-Based 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 |
|---|---|---|---|
| add spatial graph connections during training (ablation) | PCB baseline · MARS | top-1: 85.3 -> 88.6 | Helped |
| add temporal graph connections during training (ablation) | PCB baseline · MARS | top-1: 85.3 -> 88.7 | Helped |
| add both spatial and temporal graph connections (ablation) | PCB baseline · MARS | mAP: 84.6 -> 86.5 | Helped |
| add both spatial and temporal graph connections (ablation) | PCB baseline · PoseTrackReID | top-1: 77.9 -> 79.2 | Helped |
| train longer without graphical-branch assistance (ablation) | PCB baseline · MARS | top-1: 85.3 -> 84.7 | Hurt |
| include head keypoints in the graph (ablation) | KMPNet (PCB) · MARS | top-1: 89.7 -> 88.2 | Hurt |
| include head keypoints in the graph (ablation) | KMPNet (PCB) · PoseTrackReID | mAP: 82.7 -> 82.7 | No clear effect |
| use graphical-branch embedding instead of CNN embedding (ablation) | KMPNet (PCB) · MARS | top-1: 89.7 -> 61.7 | Hurt |
| use graphical-branch embedding instead of CNN embedding (ablation) | KMPNet (PCB) · MARS | mAP: 86.5 -> 57.3 | Hurt |
| apply keypoint message passing to MGH | MGH · MARS | top-1: +2.0 pp. | Helped |
| apply keypoint message passing to MGH | MGH · MARS | mAP: +0.8 pp. | Helped |
Closest papers by meaning
- Spatial-Temporal Graph Convolutional Network for Video-Based Person Re-Identification (CVPR 2020 · similarity 0.91)
- See the Forest for the Trees: Joint Spatial and Temporal Recurrent Neural Networks for Video-Based Person Re-Identification (CVPR 2017 · similarity 0.90)
- Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning (CVPR 2018 · similarity 0.89)
- Recurrent Convolutional Network for Video-Based Person Re-Identification (CVPR 2016 · similarity 0.89)
- Top-Push Video-Based Person Re-Identification (CVPR 2016 · similarity 0.89)
- Spatial-Temporal Correlation and Topology Learning for Person Re-Identification in Videos (CVPR 2021 · 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.