Research map / 3D vision / Depth and stereo
MMA: Multi-Camera Based Global Motion Averaging
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 multi-camera rotation averaging with fixed internal camera-pose constraints | Global SfM on KITTI data02 · KITTI data02 | median rotation error: 3.0 -> 1.8 | Helped |
| Use multi-camera translation averaging with fixed internal camera-pose constraints | Global SfM on KITTI data02 · KITTI data02 | median position error: 104.2 -> 7.0 | Helped |
| Use multi-camera translation averaging with fixed internal camera-pose constraints | Global SfM on KITTI data09 · KITTI data09 | median position error: 20.7 -> 1.7 | Helped |
| Select only the top-K connected edges per image instead of using all view-graph edges (ablation) | MTA translation averaging · Five datasets containing loops | camera position error | Helped |
| Increase selected edges from K = 4 to K = 8 (ablation) | MTA translation averaging · Five datasets containing loops | median and mean position error | Helped |
| Select more than K = 8 edges per image (ablation) | MTA translation averaging · Five datasets containing loops | camera position error | Hurt |
Closest papers by meaning
- Very Large-Scale Global SfM by Distributed Motion Averaging (CVPR 2018 · similarity 0.89)
- Leveraging Camera Triplets for Efficient and Accurate Structure-from-Motion (CVPR 2024 · similarity 0.83)
- Learning to Filter Outlier Edges in Global SfM (CVPR 2025 · similarity 0.83)
- PanoPose: Self-supervised Relative Pose Estimation for Panoramic Images (CVPR 2024 · similarity 0.82)
- AnyMap: Learning a General Camera Model for Structure-from-Motion with Unknown Distortion in Dynamic Scenes (CVPR 2025 · similarity 0.82)
- DeMo: Deep Motion Field Consensus with Learnable Kernels for Two-view Correspondence Learning (AAAI 2025 · similarity 0.81)
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.