Research map / Detection and segmentation / CNN features
High-Throughput Event-Based Feature Detection and Tracking on an Embedded CPU
Ethan Elms, Yasir Latif, Tat-Jun Chin
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 activity slicer instead of 15 ms fixed-time slicing (ablation) | SPEEDTrack · EDS tracking (embedded) | SFA: 0.401 -> 0.558 | Improved |
| add activity slicer instead of 15 ms fixed-time slicing (ablation) | SPEEDTrack · Embedded device | Latency (us): 71.2 -> 49.7 | Improved |
| add activity slicer instead of 15 ms fixed-time slicing (ablation) | SPEEDTrack · Embedded device | Throughput (Mev/s): 4.01 -> 4.92 | Improved |
| add Kalman filter association instead of nearest tracklet (ablation) | SPEEDTrack · EDS tracking (embedded) | SFA: 0.438 -> 0.558 | Improved |
| add Kalman filter association instead of nearest tracklet (ablation) | SPEEDTrack · Embedded device | Throughput (Mev/s): 5.02 -> 4.92 | Hurt |
| add Kalman filter association instead of nearest tracklet (ablation) | SPEEDTrack · Embedded device | Latency (us): 45.1 -> 49.7 | Hurt |
| add ignore mask instead of recomputing Harris score (ablation) | SPEEDTrack · Embedded device | Throughput (Mev/s): 3.52 -> 4.92 | Improved |
| add ignore mask instead of recomputing Harris score (ablation) | SPEEDTrack · EDS tracking (embedded) | SFA: 0.491 -> 0.558 | Improved |
| add ignore mask instead of recomputing Harris score (ablation) | SPEEDTrack · Embedded device | Latency (us): 83.2 -> 49.7 | Improved |
| add task-graph concurrency instead of thread-based concurrency (ablation) | SPEEDTrack · Embedded device | Throughput (Mev/s): 2.97 -> 4.92 | Improved |
| add task-graph concurrency instead of thread-based concurrency (ablation) | SPEEDTrack · Embedded device | Latency (us): 96.4 -> 49.7 | Improved |
| add task-graph concurrency instead of thread-based concurrency (ablation) | SPEEDTrack · EDS tracking (embedded) | SFA: 0.558 -> 0.558 | No clear effect |
Closest papers by meaning
- Learning to Detect Objects with a 1 Megapixel Event Camera (NeurIPS 2020 · similarity 0.88)
- EventPillars: Pillar-based Efficient Representations for Event Data (AAAI 2025 · similarity 0.88)
- Data-Driven Feature Tracking for Event Cameras With and Without Frames (TPAMI 2025 · similarity 0.88)
- Event Transformer$^+$. A Multi-Purpose Solution for Efficient Event Data Processing (TPAMI 2023 · similarity 0.87)
- ETAP: Event-based Tracking of Any Point (CVPR 2025 · similarity 0.87)
- FlexEvent: Towards Flexible Event-Frame Object Detection at Varying Operational Frequencies (NeurIPS 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.
