Research map / Video / Video segmentation and tracking
Un-EVIMO: Unsupervised Event-based Independent Motion Segmentation
Ziyun Wang, Jinyuan Guo, Kostas Daniilidis
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 |
|---|---|---|---|
| fine-tune E-RAFT flow on image-based RAFT flow before pseudo-labeling (ablation) | Un-EVIMO · EVIMO Table | event-masked IoU: 32±23 -> 50±21 | Improved |
| fine-tune E-RAFT flow on image-based RAFT flow before pseudo-labeling (ablation) | Un-EVIMO · EVIMO Fast | event-masked IoU: 27±23 -> 44±21 | Improved |
| use input depth in pseudo-label generation instead of 12-DOF parametric flow (ablation) | Un-EVIMO · EVIMO Table | event-masked IoU: 47±24 -> 50±21 | Improved |
| use input depth instead of 12-DOF parametric flow (ablation) | Un-EVIMO · EVIMO Floor | event-masked IoU: 56±18 -> 56±15 | No clear effect |
| use parametric flow models without depth instead of EMSGC (ablation) | Un-EVIMO pseudo-labels · EVIMO | event-masked IoU | Improved |
| fine-tune flow network on image-based flow vs pretrained E-RAFT | flow network · EVIMO Fast | EPE: 20.471 -> 5.331 | Improved |
| replace per-slice optimization (EMSGC) with a feed-forward network | IMO segmentation · EVIMO | total time (ms): 9529.42 -> 6.57 | Improved |
| use unsupervised pseudo-labels instead of GT masks | Un-EVIMO · EVIMO Table | event-masked IoU: 79±6 (supervised EVIMO) vs 50±21 | Hurt |
| frame input under synthetic motion blur instead of events | supervised video network · EVIMO Table | event-masked IoU: 66±23 -> 24±25 | Hurt |
Closest papers by meaning
- Event-Based Motion Segmentation by Motion Compensation (ICCV 2019 · similarity 0.88)
- Unsupervised 3d Motion Estimation Using Event Camera (CVPR 2026 · similarity 0.87)
- Single Image Optical Flow Estimation With an Event Camera (CVPR 2020 · similarity 0.86)
- Event-based Optical Flow Using Spatio-temporal Registration (TPAMI 2026 · similarity 0.86)
- TimeTracker: Event-based Continuous Point Tracking for Video Frame Interpolation with Non-linear Motion (CVPR 2025 · similarity 0.85)
- PASS: Path-selective State Space Model for Event-based Recognition (NeurIPS 2025 · similarity 0.85)
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.
