Research map / Video / Video segmentation and tracking
EMoTive: Event-guided Trajectory Modeling for 3D Motion Estimation
Zengyu Wan, Wei Zhai, Yang Cao, Zhengjun Zha
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 temporal gradient regularization Lt (ablation) | EMoTive · CarlaEvent3D | optical flow EPE: 3.148 -> 3.118 | Improved |
| add temporal gradient regularization Lt (ablation) | EMoTive · CarlaEvent3D | motion in depth log-mid: 139.409 -> 138.585 | Improved |
| use event-guided non-uniform trajectory instead of uniform B-spline (ablation) | EMoTive · CarlaEvent3D | optical flow EPE and log-mid: 4.6% (optical flow) and 6.9% (depth motion) | Improved |
| use event-guided non-uniform trajectory instead of uniform B-spline (ablation) | EMoTive · DSEC | optical flow EPE: 0.497 -> 0.439 | Improved |
| increase temporal bins of Event Kymograph beyond 120 (ablation) | EMoTive · CarlaEvent3D | EPE / log-mid: optimal 120 bins / 100ms | Hurt |
| use finer temporal resolution up to 120 bins (ablation) | EMoTive · CarlaEvent3D | EPE / log-mid: 60 to 120 bins | Improved |
| raise curve degree beyond 3 or lower it (ablation) | EMoTive · CarlaEvent3D | EPE / log-mid: 3rd-order optimal | Mixed |
| set control points Nc=5 vs 4 (ablation) | EMoTive · CarlaEvent3D | optical flow EPE: 0.024px decrease from Nc = 4 to 3.118px | Improved |
| set control points Nc=5 vs 6 (ablation) | EMoTive · CarlaEvent3D | log-mid: 1.6% improvement over Nc = 6 (138.585) | Improved |
| use event-driven trajectory modeling instead of spatial-correlation methods | EMoTive vs image-based methods · CarlaEvent3D | optical flow EPE: 43.5% in optical flow EPE, and 33.6% in motion in depth log-mid | Improved |
| apply image-based spatial-correlation methods to event voxels | Expansion, TPCV, ScaleFlow, Scale++ · CarlaEvent3D | optical flow EPE: 5.274 - 1.963 EPE increase | Hurt |
| EMoTive compared with ScaleFlow in size and speed | EMoTive · CarlaEvent3D | params / inference time: 5.61M vs 10.7M parameters; 40 ms, 52.9% speed improvement | Improved |
Closest papers by meaning
- Unsupervised 3d Motion Estimation Using Event Camera (CVPR 2026 · similarity 0.90)
- EF-3DGS: Event-Aided Free-Trajectory 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.89)
- E-MOTION: A Dataset for Event-Based Scene Flow Estimation with Independent Moving Objects (ECCV 2026 · similarity 0.89)
- TimeTracker: Event-based Continuous Point Tracking for Video Frame Interpolation with Non-linear Motion (CVPR 2025 · similarity 0.87)
- E-Motion: Future Motion Simulation via Event Sequence Diffusion (NeurIPS 2024 · similarity 0.87)
- Know Where You Are From: Event-Based Segmentation via Spatio-Temporal Propagation (AAAI 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.
