Research map / Detection and segmentation / Tracking
Learning Non-Uniform Hypergraph for Multi-Object Tracking
Longyin Wen, Dawei Du, Shengkun Li, Xiao Bian, Siwei Lyu
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 |
|---|---|---|---|
| Use a non-uniform hypergraph with learned weights and maximum degree 4 | NT · Multi-pedestrian tracking dataset | MOTA: 66.2 -> 68.9 | Improved |
| Use maximum hypergraph degree 2 instead of 4 (ablation) | NT · Multi-pedestrian tracking dataset | MOTA: 68.9 -> 67.5 | Hurt |
| Use maximum hypergraph degree 3 instead of 4 (ablation) | NT · Multi-pedestrian tracking dataset | MOTA: 68.9 -> 68.8 | Hurt |
| Use maximum hypergraph degree 5 instead of 4 (ablation) | NT · Multi-pedestrian tracking dataset | MOTA: 68.9 -> 68.5 | Hurt |
| Add degree-3 hyperedges (ablation) | NT · Multi-pedestrian tracking dataset | MOTA: 68.4 -> 68.9 | Improved |
| Add degree-3 and degree-4 hyperedges (ablation) | NT · Multi-pedestrian tracking dataset | MOTA: 67.6 -> 68.5 | Improved |
| Learn hyperedge-degree weights with SSVM rather than using equal weights (ablation) | NT · Multi-pedestrian tracking dataset | MOTA: 67.1 -> 67.5; 67.5 -> 68.8; 67.4 -> 68.9; 67.1 -> 68.5 | Improved |
| Use non-uniform hypergraph tracking | NT · SubwayFace | MOTA: 45.8 -> 53.1 | Improved |
Closest papers by meaning
- Exploiting Hierarchical Dense Structures on Hypergraphs for Multi-Object Tracking (TPAMI 2016 · similarity 0.89)
- Learnable Graph Matching: Incorporating Graph Partitioning With Deep Feature Learning for Multiple Object Tracking (CVPR 2021 · similarity 0.87)
- ReST: A Reconfigurable Spatial-Temporal Graph Model for Multi-Camera Multi-Object Tracking (ICCV 2023 · similarity 0.87)
- Focusing on Tracks for Online Multi-Object Tracking (CVPR 2025 · similarity 0.87)
- Probabilistic Tracklet Scoring and Inpainting for Multiple Object Tracking (CVPR 2021 · similarity 0.86)
- Multi-Scene Generalized Trajectory Global Graph Solver with Composite Nodes for Multiple Object Tracking (AAAI 2024 · similarity 0.86)
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.
