Research map / 3D vision / Depth and stereo
Progressive Structure from Motion
Alex Locher, Michal Havlena, Luc Van Gool
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 progressive clustering pipeline instead of incremental Theia (progressive mode) | Incremental SfM · Stanford | reconstructed images: 51 -> 3,165 | Improved |
| Use progressive clustering pipeline instead of incremental Theia (progressive mode) | Incremental SfM · Quad | reconstructed images: 527 -> 3,894 | Improved |
| Use progressive pipeline on shuffled TempleOfHeaven images | Incremental SfM · TempleOfHeaven (shuffled order) | model coverage | Improved |
| Use progressive pipeline in progressive mode vs batch mode for Theia runtime | Proposed · Stanford | runtime (s): 34,749 -> 13,276 | Improved |
| Use progressive pipeline instead of incremental Theia (progressive) | Incremental SfM · Quad | runtime (s): 215,064 -> 25,713 | Improved |
| Use progressive pipeline instead of HSfM batch | HSfM · Stanford | reconstructed images: 2,140 -> 3,165 | Improved |
| Use progressive pipeline instead of HSfM batch | HSfM · Quad | reconstructed images: 4,832 -> 3,894 | Improved |
| Use progressive pipeline instead of incremental Theia in progressive mode | Incremental SfM · Arc De Triomphe | runtime (s): 3,516 -> 377 | Improved |
| Use progressive pipeline instead of incremental Theia in progressive mode | Incremental SfM · BigBen | reconstructed images: 394 -> 285 | Hurt |
| Interleave images of a periodic symmetric ordering (very unfortunate order) | Proposed · TempleOfHeaven | recovery: clusters separated after 36th image | Improved |
Closest papers by meaning
- Structure-From-Motion Revisited (CVPR 2016 · similarity 0.91)
- Structure-from-Motion with a Non-Parametric Camera Model (CVPR 2025 · similarity 0.86)
- VGGSfM: Visual Geometry Grounded Deep Structure From Motion (CVPR 2024 · similarity 0.85)
- Non-Rigid Structure-from-Motion: Temporally-Smooth Procrustean Alignment and Spatially-Variant Deformation Modeling (CVPR 2024 · similarity 0.85)
- ReMatching Dynamic Reconstruction Flow (ICLR 2025 · similarity 0.84)
- Light3R-SfM: Towards Feed-forward Structure-from-Motion (CVPR 2025 · similarity 0.84)
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.
