Research map / 3D vision / Depth and stereo
Motion and Structure from Event-based Normal Flow
Zhongyang Ren, Bangyan Liao, Delei Kong, Jinghang Li, Peidong Liu, Laurent Kneip, Guillermo Gallego, Yi Zhou
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 the proposed normal-flow constraint instead of naive substitution of normal flow for full flow (ablation) | 2D registration toy example · toy registration | registration bias | Improved |
| replace CMax with the proposed linear solver for angular velocity | angular velocity estimation · shapes_rotation | e_w (deg/s): 15.60 -> 15.36 | Improved |
| replace Pro-STR with the proposed linear solver for angular velocity | angular velocity estimation · shapes_rotation | e_w (deg/s): 90.61 -> 15.36 | Improved |
| initialize CMax with the linear solver (ablation) | CMax · ground_rotation | e_w (deg/s): 9.48 -> 2.85 | Improved |
| initialize CMax with the linear solver (ablation) | CMax · patterns_rotation | e_w (deg/s): 5.95 -> 0.35 | Improved |
| initialize Pro-STR with the linear solver (ablation) | Pro-STR · board_slow differential homography | F-Norm: 2.07 -> 0.19 | Improved |
| initialize CMax with the linear solver (ablation) | CMax · mountain_normal 6-DoF | e_w (deg/s): 11.01 -> 3.84 | Improved |
| refine the linear solution with nonlinear solvers (Ours+CMax / Ours+Pro-STR) (ablation) | linear solver · corner_slow 6-DoF | e_w (deg/s): 1.31 -> 2.19 (CMax); 1.31 -> 1.06 (Pro-STR) | Mixed |
| use continuous-time nonlinear solver instead of constant-motion CMax and Pro-STR | angular velocity under sudden speed changes · shapes_rotation segments | agreement with ground truth | Improved |
Closest papers by meaning
- A 5-Point Minimal Solver for Event Camera Relative Motion Estimation (ICCV 2023 · similarity 0.90)
- A Linear N-Point Solver for Structure and Motion from Asynchronous Tracks (ICCV 2025 · similarity 0.89)
- Learning Normal Flow Directly From Events (ICCV 2025 · similarity 0.89)
- Event-based Optical Flow Using Spatio-temporal Registration (TPAMI 2026 · similarity 0.89)
- An N-Point Linear Solver for Line and Motion Estimation with Event Cameras (CVPR 2024 · similarity 0.89)
- The Spatio-Temporal Poisson Point Process: A Simple Model for the Alignment of Event Camera Data (ICCV 2021 · similarity 0.88)
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.
