Research map / 3D vision / Depth and stereo
Hybrid Scene Compression for Visual Localization
Federico Camposeco, Andrea Cohen, Marc Pollefeys, Torsten Sattler
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 |
|---|---|---|---|
| Add image subdivisions to point selection (ablation) | Hybrid scene compression on Dubrovnik · Dubrovnik, 1.5% memory | registered images: 95.7% -> 96.0% | Improved |
| Use modified RANSAC, including compressed-point matches for verification (ablation) | Hybrid scene compression on Dubrovnik · Dubrovnik, 1.5% memory | registered images: 95.7% -> 97.4% | Improved |
| Use both image subdivisions and modified RANSAC (ablation) | Hybrid scene compression on Dubrovnik · Dubrovnik, 1.5% memory | median position error: 2.09 -> 1.97 | Improved |
| Combine full-descriptor points with additional quantized-descriptor points | Hybrid scene compression · Six real-world datasets | registration rate | Improved |
| Use image subdivisions for more even point coverage | Hybrid scene compression · Six real-world datasets | registration rate | Improved |
| Use quantized word occupancy to approximate descriptor distinctiveness during compression | Hybrid scene compression · Six real-world datasets | compression time: 27.9 vs 50.4 | Improved |
| Use co-visibility-guided sampling in RANSAC | Modified RANSAC · Visual localization | localization rate | Improved |
Closest papers by meaning
- Simple but Effective Triplet-Based Compression Strategies for Compact Visual Localization (CVPR 2026 · similarity 0.92)
- SceneSqueezer: Learning To Compress Scene for Camera Relocalization (CVPR 2022 · similarity 0.90)
- MAD-DR: Map Compression for Visual Localization with Matchness Aware Descriptor Dimension Reduction (ECCV 2024 · similarity 0.88)
- Revisiting Monocular SLAM with Spatio-Temporal Scene Modeling (CVPR 2026 · similarity 0.87)
- Learning to Produce Semi-dense Correspondences for Visual Localization (CVPR 2024 · similarity 0.84)
- Gaussian Splatting Feature Fields for (Privacy-Preserving) Visual Localization (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.
