Research map / 3D vision / Depth and stereo
A Linear N-Point Solver for Structure and Motion from Asynchronous Tracks
Hang Su, Yunlong Feng, Daniel Gehrig, Panfeng Jiang, Ling Gao, Xavier Lagorce, Laurent Kneip
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 rolling shutter timestamp correction (ablation) | Ours · TUM rolling shutter Seq 4 | mean velocity error: 28.1 -> 27.5 | Improved |
| add rolling shutter timestamp correction (ablation) | Ours · TUM rolling shutter Seq 5 | mean velocity error: 27.0 -> 24.7 | Improved |
| use global shutter tracks with point solver instead of eventail on events | Ours · desk-normal | mean velocity error: 22.7 -> 15.1 | Improved |
| point solver instead of eventail on rolling shutter edges | Ours · TUM rolling shutter Seq 4 | mean velocity error: 43.8 -> 27.5 | Improved |
| point solver instead of eventail on event tracks | Ours + E · boxes_translation | mean velocity error: 34.3 -> 12.6 | Improved |
| add global shutter tracks to event tracks | Ours + E · shapes_translation | mean velocity error: 16.8 -> 14.4 | Improved |
| add global shutter tracks to event tracks | Ours + E · boxes_translation | mean velocity error: 12.6 -> 10.3 | Improved |
| add events to global shutter tracks | Ours · shapes_translation | mean velocity error: 17.1 -> 14.4 | Improved |
| restrict to subset with track inlier ratio above 0.9 | Ours · desk-normal | mean velocity error: 15.1 -> 10.2 | Improved |
| increase number of tracks from 3 to 30 | Ours · simulation | velocity error | Improved |
| increase observations per track under fixed time window | Ours · simulation | velocity error | No clear effect |
| use larger time window (longer tracks) | Ours · simulation | velocity error | Improved |
Closest papers by meaning
- Motion and Structure from Event-based Normal Flow (ECCV 2024 · similarity 0.89)
- A 5-Point Minimal Solver for Event Camera Relative Motion Estimation (ICCV 2023 · similarity 0.89)
- An N-Point Linear Solver for Line and Motion Estimation with Event Cameras (CVPR 2024 · similarity 0.88)
- The Spatio-Temporal Poisson Point Process: A Simple Model for the Alignment of Event Camera Data (ICCV 2021 · similarity 0.88)
- Event-aided Dense and Continuous Point Tracking: Everywhere and Anytime (ICCV 2025 · similarity 0.87)
- Track-On: Transformer-based Online Point Tracking with Memory (ICLR 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.
