Research map / 3D vision / Depth and stereo
Learning Visual Motion Segmentation Using Event Surfaces
Anton Mitrokhin, Zhiyuan Hua, Cornelia Fermuller, Yiannis Aloimonos
Check 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 |
|---|---|---|---|
| increase event-slice width from 0.02 s to 0.3 s (ablation) | GConv · EV-IMO boxes | mIoU (%): 60±18 -> 84±9 | Improved |
| increase event-slice width from 0.02 s to 0.3 s (ablation) | GConv · EV-IMO floor | mIoU (%): 55±19 -> 80±9 | Improved |
| increase event-slice width from 0.02 s to 0.3 s (ablation) | GConv · EV-IMO wall | mIoU (%): 51±16 -> 85±8 | Improved |
| increase event-slice width from 0.02 s to 0.3 s (ablation) | GConv · EV-IMO table | mIoU (%): 57±14 -> 87±7 | Improved |
| increase event-slice width from 0.02 s to 0.3 s (ablation) | GConv · EV-IMO fast | mIoU (%): 39±19 -> 77±10 | Improved |
| subsample events by factor u=2 instead of using the full cloud (ablation) | GConv · EV-IMO boxes | mIoU (%): 84±9 -> 85±5 | No clear effect |
| subsample events by factor u=2 instead of using the full cloud (ablation) | GConv · EV-IMO floor | mIoU (%): 80±9 -> 81±12 | No clear effect |
| subsample events by factor u=2 instead of using the full cloud (ablation) | GConv · EV-IMO wall | mIoU (%): 85±8 -> 83±4 | No clear effect |
| subsample events by factor u=2 instead of using the full cloud (ablation) | GConv · EV-IMO table | mIoU (%): 87±7 -> 85±11 | No clear effect |
| subsample events by factor u=2 instead of using the full cloud (ablation) | GConv · EV-IMO fast | mIoU (%): 77±10 -> 74±19 | No clear effect |
| use temporal augmentation by randomly scaling event time (ablation) | GConv · EV-IMO fast validation after training on boxes | mIoU (%) | Improved |
| reduce maximum edge radius from 10 to 7 or 5 pixels (ablation) | GConv · EV-IMO boxes validation | mIoU (%) | Hurt |
Closest papers by meaning
- Event-Based Motion Segmentation by Motion Compensation (ICCV 2019 · similarity 0.92)
- Instance-Level Moving Object Segmentation from a Single Image with Events (IJCV 2025 · similarity 0.90)
- Know Where You Are From: Event-Based Segmentation via Spatio-Temporal Propagation (AAAI 2025 · similarity 0.88)
- Learning Segmentation from Point Trajectories (NeurIPS 2024 · similarity 0.88)
- Efficient Event-Based Semantic Segmentation via Exploiting Frame-Event Fusion: A Hybrid Neural Network Approach (AAAI 2025 · similarity 0.87)
- PASS: Path-selective State Space Model for Event-based Recognition (NeurIPS 2025 · similarity 0.87)
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.
