Research map / Video / Video segmentation and tracking
Data-Driven Feature Tracking for Event Cameras
Nico Messikommer, Carter Fang, Mathias Gehrig, Davide Scaramuzza
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 the learned tracker instead of EKLT | Event-based feature tracking · EDS | Expected FA: 0.205 -> 0.451 | Improved |
| Use the learned tracker instead of EKLT | Event-based feature tracking · EC | Expected FA: 0.775 -> 0.787 | Improved |
| Fine-tune the tracker with pose supervision | Ours (zero-shot) · EDS | Expected FA: 0.451 -> 0.472 | Improved |
| Fine-tune the tracker with pose supervision | Ours (zero-shot) · EC | Expected FA: 0.787 -> 0.818 | Improved |
| Add training-time augmentation (ablation) | Reference Model · EDS | Expected FA: 0.178 -> 0.383 | Improved |
| Add training-time augmentation (ablation) | Reference Model · EC | Expected FA: 0.599 -> 0.787 | Improved |
| Add recurrence to the feature encoder (ablation) | Reference Model · EDS | Expected FA: 0.301 -> 0.383 | Improved |
| Add recurrence to the feature encoder (ablation) | Reference Model · EC | Expected FA: 0.606 -> 0.787 | Improved |
| Add the correlation map (ablation) | Reference Model · EDS | Expected FA: 0.341 -> 0.383 | Improved |
| Add the correlation map (ablation) | Reference Model · EC | Expected FA: 0.684 -> 0.787 | Improved |
| Add frame attention to the reference model (ablation) | Reference Model · EDS | Expected FA: 0.383 -> 0.451 | Improved |
| Add frame attention to the reference model (ablation) | Reference Model · EC | Expected FA: 0.787 -> 0.787 | No clear effect |
Closest papers by meaning
- Data-Driven Feature Tracking for Event Cameras With and Without Frames (TPAMI 2025 · similarity 0.95)
- Asynchronous, Photometric Feature Tracking using Events and Frames (ECCV 2018 · similarity 0.90)
- BlinkTrack: Feature Tracking over 80 FPS via Events and Images (ICCV 2025 · similarity 0.89)
- EKLT: Asynchronous Photometric Feature Tracking Using Events and Frames (IJCV 2019 · similarity 0.89)
- ETAP: Event-based Tracking of Any Point (CVPR 2025 · similarity 0.88)
- Event Stream-based Visual Object Tracking: A High-Resolution Benchmark Dataset and A Novel Baseline (CVPR 2024 · 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.
