Research map / 3D vision / Depth and stereo
A Dataset for Benchmarking Image-Based Localization
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 COV features instead of SIFT features | Retrieval-based localization · Proposed dataset | registration rate: 59.2% -> 74.2% | Helped |
| Add RootSIFT to the COV retrieval baseline | Retrieval-based localization · Proposed dataset | registration rate: 74.2% -> 75.4% | Helped |
| Add AQE (ablation) | COV + RootSIFT retrieval · Proposed dataset | registration rate: 75.4% -> 75.4% | No clear effect |
| Add SPAUG (ablation) | COV + RootSIFT retrieval · Proposed dataset | registration rate: 75.4% -> 75.8% | Helped |
| Add selective voting (ablation) | COV + RootSIFT retrieval · Proposed dataset | registration rate: 75.4% -> 77.8% | Helped |
| Add virtual view synthesis (ablation) | RootSIFT retrieval · Proposed dataset | registration rate: +9.4% | Helped |
| Add reranking to RootSIFT + VVS (ablation) | Retrieval-based localization · Proposed dataset | registration rate: 84.8% -> 83.9% | Hurt |
| Use direct matching instead of the retrieval-based RootSIFT + VVS configuration | Image-based localization · Proposed dataset | registration rate: 84.8% -> 83.3% | Hurt |
| Use unknown camera intrinsics and a 6-point-pose solver instead of pre-calibrated intrinsics (ablation) | COV + VVS + RERANK · Proposed dataset | registration rate: decreases by 16.4% | Hurt |
| Replace the LiDAR point cloud with a simulated, sparser SfM point cloud (ablation) | COV retrieval baseline · Proposed dataset | registration rate: drops by 15% | Hurt |
Closest papers by meaning
- Large-Scale Localization Datasets in Crowded Indoor Spaces (CVPR 2021 · similarity 0.92)
- 360Loc: A Dataset and Benchmark for Omnidirectional Visual Localization with Cross-device Queries (CVPR 2024 · similarity 0.88)
- Rooms from Motion: Un-posed Indoor 3D Object Detection as Localization and Mapping (NeurIPS 2025 · similarity 0.87)
- RELOCATE: A Simple Training-Free Baseline for Visual Query Localization Using Region-Based Representations (CVPR 2025 · similarity 0.87)
- Learning to Produce Semi-dense Correspondences for Visual Localization (CVPR 2024 · similarity 0.87)
- LiDAR-Net: A Real-scanned 3D Point Cloud Dataset for Indoor Scenes (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.