Research map / Video / Video segmentation and tracking
Event-Based Motion Segmentation by Motion Compensation
Timo Stoffregen, Guillermo Gallego, Tom Drummond, Lindsay Kleeman, 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 |
|---|---|---|---|
| add GPU execution | Ours (Alg. 1) · Throughput, Nl=2 | throughput (kevents/s): 239.86 -> 3963.20 | Improved |
| increase number of clusters from 2 to 50 (ablation) | Ours (Alg. 1) CPU · Throughput | throughput (kevents/s): 239.86 -> 12.62 | Hurt |
| increase number of clusters from 2 to 50 (ablation) | Ours (Alg. 1) GPU · Throughput | throughput (kevents/s): 3963.20 -> 113.78 | Hurt |
| increase relative displacement between objects (ablation) | Ours (Alg. 1) · Simulated pebbles | per-event accuracy: 90 % at 4 pixels | Improved |
| use 5, 10 or 20 optical-flow clusters (ablation) | Ours (Alg. 1) · slider depth sequence | qualitative segmentation | No clear effect |
| add five rotation clusters to five optical-flow clusters (ablation) | Ours (Alg. 1) · slider depth sequence | qualitative segmentation | No clear effect |
| use only rotation motion models (ablation) | Ours (Alg. 1) · slider depth sequence | qualitative segmentation | Hurt |
| replace SOFAS greedy fitting with joint per-event segmentation | Ours (Alg. 1) · EED Multiple objects | success rate (%): 46.15 -> 96.77 | Improved |
| replace Mitrokhin with joint per-event segmentation | Ours (Alg. 1) · EED Lighting variation | success rate (%): 84.52 -> 80.51 | Hurt |
| replace Mitrokhin with joint per-event segmentation | Ours (Alg. 1) · EED What is Background? | success rate (%): 89.21 -> 100.00 | Improved |
Closest papers by meaning
- Learning Visual Motion Segmentation Using Event Surfaces (CVPR 2020 · similarity 0.92)
- Continuous-Time Object Segmentation Using High Temporal Resolution Event Camera (TPAMI 2024 · similarity 0.92)
- Instance-Level Moving Object Segmentation from a Single Image with Events (IJCV 2025 · similarity 0.92)
- Event-Based Vision: A Survey (TPAMI 2020 · similarity 0.91)
- ESS: Learning Event-Based Semantic Segmentation from Still Images (ECCV 2022 · similarity 0.90)
- Event-assisted Low-Light Video Object Segmentation (CVPR 2024 · 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.
