Research map / 3D vision / Depth and stereo
Improved Structure from Motion Using Fiducial Marker Matching
Joseph DeGol, Timothy Bretl, Derek Hoiem
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 marker informed resectioning (ablation) | marker-assisted SfM · Marker SfM dataset | avg % images localized: 68% -> 98% | Improved |
| add marker informed matching (pair filtering) (ablation) | marker-assisted SfM · Marker SfM dataset | avg % images localized: 50% -> 98% | Improved |
| add marker scale and orthogonality constraints to bundle adjustment (ablation) | marker-assisted SfM · Marker SfM dataset | reconstruction quality | No clear effect |
| use marker detections as tracks in OpenSfM | OpenSfM · Marker SfM dataset | reconstruction quality | No clear effect |
| leave markers visible as texture in OpenSfM instead of masking them | OpenSfM · Marker SfM dataset | reconstruction quality | Hurt |
| use markers to limit image pairs for matching | OpenSfM · MUF Floor2 | matching time (s): 5596 seconds and the other OpenSfM approaches took 5-6 times longer | Improved |
| increase average marker detections per image | marker-assisted SfM · Marker SfM dataset (7 sets) | % images localized: AMD of 6 gives 90+% localized | Improved |
Closest papers by meaning
- Leveraging Camera Triplets for Efficient and Accurate Structure-from-Motion (CVPR 2024 · similarity 0.84)
- Detector-Free Structure from Motion (CVPR 2024 · similarity 0.84)
- MP-SfM: Monocular Surface Priors for Robust Structure-from-Motion (CVPR 2025 · similarity 0.84)
- Shading Meets Motion: Self-supervised Indoor 3D Reconstruction Via Simultaneous Shape-from-Shading and Structure-from-Motion (CVPR 2025 · similarity 0.84)
- VGGSfM: Visual Geometry Grounded Deep Structure From Motion (CVPR 2024 · similarity 0.83)
- Light3R-SfM: Towards Feed-forward Structure-from-Motion (CVPR 2025 · similarity 0.83)
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.
