Research map / Detection and segmentation / Tracking
Learning Graph-embedded Key-event Back-tracing for Object Tracking in Event Clouds
Zhiyu Zhu, Junhui Hou, Xianqiang 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 |
|---|---|---|---|
| add density-insensitive event downsampling instead of random sampling (ablation) | Ours · FE108 ALL | RSR: 50.1 -> 54.9 | Improved |
| add density-insensitive event downsampling instead of random sampling (ablation) | Ours · FE108 ALL | RPR: 77.5 -> 85.9 | Improved |
| add motion-aware target likelihood prediction with noise-injection training strategy (ablation) | Ours · FE108 ALL | RSR: 51.7 -> 53.9 | Improved |
| add motion-aware target likelihood prediction without the training strategy (ablation) | Ours · FE108 ALL | RSR: 52.2 -> 51.7 | Hurt |
| add flow estimation module (ablation) | Ours · FE108 ALL | RSR: 53.9 -> 54.9 | Improved |
| add online template updating strategy (ablation) | Ours · FE108 ALL | RSR: 54.5 -> 54.9 | Improved |
| use raw event input instead of event frames, versus E-TransT EF-3 | Ours · FE108 ALL | RSR: 52.4 -> 54.9 | Improved |
| increase E-TransT input from EF-1 to EF-3 (higher temporal resolution) | E-TransT · FE108 ALL | RSR: 39.1 -> 52.4 | Improved |
| use raw event tracker instead of 3D point cloud trackers | Ours · FE108 ALL | RSR: 47.5 -> 54.9 | Improved |
| use raw event tracker instead of E-TransT on synthetic data | Ours · Evt-LaSOT | RSR: 30.3 -> 32.3 | Improved |
| use raw event tracker instead of E-TransT on synthetic data (speed) | Ours · Evt-LaSOT | FPS: 18.7 -> 25.2 | Improved |
| use raw event tracker versus E-PrDiMP in the HDR scenario | Ours · FE108 HDR | RPR: 84.6 -> 81.5 | Hurt |
Closest papers by meaning
- Event Stream-based Visual Object Tracking: A High-Resolution Benchmark Dataset and A Novel Baseline (CVPR 2024 · similarity 0.88)
- Event-aided Dense and Continuous Point Tracking: Everywhere and Anytime (ICCV 2025 · similarity 0.87)
- Rethinking Efficient and Effective Point-Based Networks for Event Camera Classification and Regression (TPAMI 2025 · similarity 0.87)
- Data-Driven Feature Tracking for Event Cameras (CVPR 2023 · similarity 0.87)
- PASS: Path-selective State Space Model for Event-based Recognition (NeurIPS 2025 · similarity 0.87)
- ETAP: Event-based Tracking of Any Point (CVPR 2025 · 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.
