Research map / 3D vision / Depth and stereo
View-graph Selection Framework for SfM
Rajvi Shah, Visesh Chari, P J Narayanan
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 |
|---|---|---|---|
| select edges with view-graph selection instead of full graph | Global SfM · Internet landmarks - Pantheon | SfM runtime: 3601 -> 1785 | Improved |
| select edges with view-graph selection instead of full graph | Global SfM · Internet landmarks - Notre Dame | SfM runtime: 1760 -> 1151 | Improved |
| select edges with view-graph selection instead of full graph | Global SfM · Internet landmarks - Pantheon | rotation error: 0.125 -> 0.098 | Improved |
| select edges with view-graph selection instead of full graph | Global SfM · Internet landmarks - St. Peters | rotation error: 0.028 -> 0.037 | Hurt |
| select edges with view-graph selection instead of full graph | Global SfM · Internet landmarks - Castle (MVS) | translation error: 7.49 -> 1.29 | Improved |
| select edges with view-graph selection instead of full graph | Incremental SfM · MVS benchmark | reconstruction accuracy | No clear effect |
| use random costs for selection (ablation) | View-graph selection · Ambiguous datasets | successful reconstruction: fails on all datasets | Hurt |
| use baseline priors based costs (ablation) | View-graph selection · Ambiguous datasets | successful reconstruction: fails on all but books | Hurt |
| add disambiguation priors to general reconstruction priors (ablation) | View-graph selection · Ambiguous datasets | successful reconstruction: fails on cup and cereal only | Improved |
| use context similarity prior alone (ablation) | View-graph selection · Ambiguous small datasets (6 lab-style) | successful reconstruction: 6 of 6 correct | Improved |
| combine context and loop priors (ablation) | View-graph selection · Ambiguous datasets | range of flow values | Improved |
| add multiple-motions prior to context and loop priors (ablation) | View-graph selection · Large ambiguous datasets | successful reconstruction | No clear effect |
Closest papers by meaning
- Learning Structure-from-Motion with Graph Attention Networks (CVPR 2024 · similarity 0.87)
- Leveraging Camera Triplets for Efficient and Accurate Structure-from-Motion (CVPR 2024 · similarity 0.86)
- Light3R-SfM: Towards Feed-forward Structure-from-Motion (CVPR 2025 · similarity 0.86)
- Dense-SfM: Structure from Motion with Dense Consistent Matching (CVPR 2025 · similarity 0.86)
- MP-SfM: Monocular Surface Priors for Robust Structure-from-Motion (CVPR 2025 · similarity 0.86)
- Learning to Filter Outlier Edges in Global SfM (CVPR 2025 · similarity 0.85)
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.
