Research map / 3D vision / Depth and stereo
Large-Scale, Metric Structure From Motion for Unordered Light Fields
Sotiris Nousias, Manolis Lourakis, Christos Bergeles
Check 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 uLF-SfM instead of COLMAP on the same LF datasets | 3D reconstruction · Octopus, House, Toycar, and Chameleon | 3D points: 1045 -> 1169; 1654 -> 1756; 9917 -> 13512; 28281 -> 35597 | Improved |
| Use uLF-SfM instead of COLMAP on central sub-aperture images | 3D reconstruction · Octopus, House, and Chameleon | Avg. reprojection error [pix]: 0.16; 0.24; 0.23; 0.21 | No clear effect |
| Use uLF-SfM instead of COLMAP, evaluating reprojection error across all sub-aperture images | 3D reconstruction · Octopus, House, Toycar, and Chameleon | Avg. reprojection error [pix]: 0.29 vs 0.20; 0.31 vs 0.26; 0.72 vs 0.17; 0.81 vs 0.19 | Mixed |
| Use uLF-SfM rather than LF-SfM | LF-SfM reconstruction · Toycar and Chameleon | registered frames: LF-SfM registered only 63 frames on “Toycar” and failed on “Chameleon”. | Improved |
| Recover poses with uLF-SfM rather than LF-SfM | LF-SfM reconstruction · Octopus and House | Transl. Difference: 0.16 ± 0.13 vs 4.39 ± 2.36; 0.13 ± 0.08 vs 3.34 ± 1.51 | Improved |
| Use uLF-SfM rather than COLMAP on the reported datasets | LF camera pose estimation · Octopus, House, Toycar, and Chameleon | pose difference from COLMAP: 0.16 ± 0.13; 0.13 ± 0.08; 0.11 ± 0.04; 0.12 ± 0.08 | No clear effect |
Closest papers by meaning
- VGGSfM: Visual Geometry Grounded Deep Structure From Motion (CVPR 2024 · similarity 0.88)
- Light3R-SfM: Towards Feed-forward Structure-from-Motion (CVPR 2025 · similarity 0.87)
- Reconstructing People, Places, and Cameras (CVPR 2025 · similarity 0.87)
- No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images (ICLR 2025 · similarity 0.86)
- Large Spatial Model: End-to-end Unposed Images to Semantic 3D (NeurIPS 2024 · similarity 0.86)
- Detector-Free Structure from Motion (CVPR 2024 · 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.
