Research map / 3D vision / Depth and stereo
Time-to-Contact Map by Joint Estimation of Up-to-Scale Inverse Depth and Global Motion using a Single Event Camera
Urbano Miguel Nunes, Laurent Udo Perrinet, Sio-Hoi Ieng
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 |
|---|---|---|---|
| use proposed Translation model (s=2) instead of ECMD for global TTC | VL dataset · VL dataset | REE avg (%): 8.92 -> 5.64 | Improved |
| use proposed method instead of ECMD for runtime per event | VL dataset · VL dataset | runtime per event avg (microsec): 50.90 -> 1.15 | Improved |
| increase neighborhood size from s=2 to s=3 for Scaling model (ablation) | Scaling model · VL dataset | REE avg (%): 8.29 -> 15.08 | Hurt |
| increase neighborhood size from s=2 to s=3 for 6-DOF model (ablation) | 6-DOF model · VL dataset | REE avg (%): 10.11 -> 6.42 | Improved |
| increase neighborhood size from s=2 to s=3 (ablation) | proposed method · VL dataset | runtime per event avg (microsec): 1.46 -> 2.36 (6-DOF) | Hurt |
| use Translation model instead of 6-DOF for divergence (ablation) | proposed method · MVSEC indoor flying1 | AREE (%): 53.65 -> 19.82 | Improved |
| use Driving motion model on non-driving indoor sequences (ablation) | proposed method · MVSEC indoor flying1 | AREE (%): 19.82 -> 153.30 | Hurt |
| add extra DOFs (Translation, 6-DOF) to Scaling motion model (ablation) | proposed method · VL dataset | REE avg (%): 8.29 -> 5.64 (Translation); 8.29 -> 10.11 (6-DOF) at s=2 | Mixed |
| use proposed method versus event-based optical flow methods | MVSEC optical flow · MVSEC | AEE (pixel/frame) | Mixed |
| use proposed Driving model for outdoor day1 optical flow instead of EV-FlowNet | MVSEC outdoor day1 · MVSEC outdoor day1 | AAE (deg): 22.88 -> 13.31 | Improved |
| transform only neighbouring events to current event timestamp, not all events to common reference | Dispersion Minimization · event collapse mitigation | stability of estimates | Improved |
Closest papers by meaning
- Event-Based Stereo Depth Estimation: A Survey (TPAMI 2025 · similarity 0.88)
- TMA: Temporal Motion Aggregation for Event-based Optical Flow (ICCV 2023 · similarity 0.87)
- Globally-Optimal Event Camera Motion Estimation (ECCV 2020 · similarity 0.87)
- Binary TTC: A Temporal Geofence for Autonomous Navigation (CVPR 2021 · similarity 0.87)
- Event Fields: Capturing Light Fields at High Speed, Resolution, and Dynamic Range (CVPR 2025 · similarity 0.85)
- EDCFlow: Exploring Temporally Dense Difference Maps for Event-based Optical Flow Estimation (CVPR 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.
