mlsanityData to 7 Oct 2026

Map-free Visual Relocalization: Metric Pose Relative to a Single Image

arXiv 2210.05494 ↗First posted 11 Oct 2022cs.CVcited by 112

Not on the map. Why?

Abstract (arXiv)

Can we relocalize in a scene represented by a single reference image? Standard visual relocalization requires hundreds of images and scale calibration to build a scene-specific 3D map. In contrast, we propose Map-free Relocalization, i.e., using only one photo of a scene to enable instant, metric scaled relocalization. Existing datasets are not suitable to benchmark map-free relocalization, due to their focus on large scenes or their limited variability. Thus, we have constructed a new dataset of 655 small places of interest, such as sculptures, murals and fountains, collected worldwide. Each place comes with a reference image to serve as a relocalization anchor, and dozens of query images with known, metric camera poses. The dataset features changing conditions, stark viewpoint changes, high variability across places, and queries with low to no visual overlap with the reference image. We identify two viable families of existing methods to provide baseline results: relative pose regression, and feature matching combined with single-image depth prediction. While these methods show reasonable performance on some favorable scenes in our dataset, map-free relocalization proves to be a challenge that requires new, innovative solutions.

Key facts. Cited by 112 papers in the map (data through 7 Oct 2026). Builds on 55 of its 87 references that are in the map. First posted on arXiv 11 Oct 2022.

Builds on

55 of 87 references.

References · 55

  1. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  2. 2021Vision Transformers for Dense Predictioncited by 1,789
  3. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,484
  4. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  5. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,264
  6. 2018On the Continuity of Rotation Representations in Neural Networkscited by 1,186
  7. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
  8. 2015NetVLAD: CNN architecture for weakly supervised place recognitioncited by 935
  9. 2016Unsupervised Monocular Depth Estimation with Left-Right Consistencycited by 819
  10. 2017Unsupervised Learning of Depth and Ego-Motion from Videocited by 778
  11. 2018MegaDepth: Learning Single-View Depth Prediction from Internet Photoscited by 731
  12. 2018From Coarse to Fine: Robust Hierarchical Localization at Large Scalecited by 556
38 more
  1. 2016Semantic Scene Completion from a Single Depth Imagecited by 468
  2. 2020DISK: Learning local features with policy gradientcited by 310
  3. 2016DSAC - Differentiable RANSAC for Camera Localizationcited by 264
  4. 2017Geometric Loss Functions for Camera Pose Regression with Deep Learningcited by 224
  5. 2020Image Matching across Wide Baselines: From Paper to Practicecited by 224
  6. 2019R2D2: Repeatable and Reliable Detector and Descriptorcited by 206
  7. 2016DeMoN: Depth and Motion Network for Learning Monocular Stereocited by 200
  8. 2017Learning to Find Good Correspondencescited by 176
  9. 2020Visual Camera Re-Localization from RGB and RGB-D Images Using DSACcited by 176
  10. 2017Geometry-Aware Learning of Maps for Camera Localizationcited by 161
  11. 2019MAGSAC++, a fast, reliable and accurate robust estimatorcited by 160
  12. 2017CNN-SLAM: Real-time dense monocular SLAM with learned depth predictioncited by 152
  13. 2017Learning Less is More - 6D Camera Localization via 3D Surface Regressioncited by 146
  14. 2016Image-based localization using LSTMs for structured feature correlationcited by 136
  15. 2019Neural-Guided RANSAC: Learning Where to Sample Model Hypothesescited by 129
  16. 2019Self-Supervised Monocular Depth Hintscited by 127
  17. 2018MAGSAC: marginalizing sample consensuscited by 122
  18. 2021Learning Multi-Scene Absolute Pose Regression with Transformerscited by 108
  19. 2018PlaneRCNN: 3D Plane Detection and Reconstruction from a Single Imagecited by 105
  20. 2019Hierarchical Scene Coordinate Classification and Regression for Visual Localizationcited by 93
  21. 2018Deep Virtual Stereo Odometry: Leveraging Deep Depth Prediction for Monocular Direct Sparse Odometrycited by 85
  22. 2018Discovery of Latent 3D Keypoints via End-to-end Geometric Reasoningcited by 79
  23. 2019CAM-Convs: Camera-Aware Multi-Scale Convolutions for Single-View Depthcited by 75
  24. 2019Expert Sample Consensus Applied to Camera Re-Localizationcited by 70
  25. 2020Robust Image Retrieval-based Visual Localization using Kapturecited by 66
  26. 2021On the Limits of Pseudo Ground Truth in Visual Camera Re-localisationcited by 63
  27. 2016Learning to Navigate the Energy Landscapecited by 62
  28. 2019To Learn or Not to Learn: Visual Localization from Essential Matricescited by 61
  29. 2016CNN Image Retrieval Learns from BoW: Unsupervised Fine-Tuning with Hard Examplescited by 58
  30. 2021Wide-Baseline Relative Camera Pose Estimation with Directional Learningcited by 49
  31. 2017Relative Camera Pose Estimation Using Convolutional Neural Networkscited by 47
  32. 2020Benchmarking Image Retrieval for Visual Localizationcited by 42
  33. 2021Extreme Rotation Estimation using Dense Correlation Volumescited by 42
  34. 2020Pseudo RGB-D for Self-Improving Monocular SLAM and Depth Predictioncited by 41
  35. 2019Reinforced Feature Points: Optimizing Feature Detection and Description for a High-Level Taskcited by 39
  36. 2021Visual Camera Re-Localization Using Graph Neural Networks and Relative Pose Supervisioncited by 37
  37. 2018RPNet: an End-to-End Network for Relative Camera Pose Estimationcited by 31
  38. 2020Beyond Controlled Environments: 3D Camera Re-Localization in Changing Indoor Scenescited by 31

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 9; 2024: 26; 2025: 36; 2026: 40.

112 papers.

Sort:

By area:

  1. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,526
  2. 2024Grounding Image Matching in 3D with MASt3Rcited by 960
  3. 2025Depth Anything 3: Recovering the Visual Space from Any Viewscited by 626
  4. 2025Continuous 3D Perception Model with Persistent Statecited by 527
  5. 20243D Reconstruction with Spatial Memorycited by 269
  6. 2024No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Imagescited by 241
  7. 2025UniDepthV2: Universal Monocular Metric Depth Estimation Made Simplercited by 170
  8. 2025MUSt3R: Multi-view Network for Stereo 3D Reconstructioncited by 139
  9. 2023Accelerated Coordinate Encoding: Learning to Relocalize in Minutes using RGB and Posescited by 107
  10. 2024SLAM3R: Real-Time Dense Scene Reconstruction from Monocular RGB Videoscited by 100
  11. 2024Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localizationcited by 81
  12. 2025STream3R: Scalable Sequential 3D Reconstruction with Causal Transformercited by 76
38 more
  1. 2025Uni3C: Unifying Precisely 3D-Enhanced Camera and Human Motion Controls for Video Generationcited by 75
  2. 2024Scene Coordinate Reconstruction: Posing of Image Collections via Incremental Learning of a Relocalizercited by 73
  3. 2025SpatialVID: A Large-Scale Video Dataset with Spatial Annotationscited by 72
  4. 2024PF3plat: Pose-Free Feed-Forward 3D Gaussian Splattingcited by 60
  5. 2025UniK3D: Universal Camera Monocular 3D Estimationcited by 60
  6. 2025WorldMirror: Universal 3D World Reconstruction with Any-Prior Promptingcited by 60
  7. 2025Dens3R: A Foundation Model for 3D Geometry Predictioncited by 44
  8. 2025RoMa v2: Harder Better Faster Denser Feature Matchingcited by 44
  9. 2024Map-Relative Pose Regression for Visual Re-Localizationcited by 42
  10. 2026NeoVerse: Enhancing 4D World Model with in-the-wild Monocular Videoscited by 39
  11. 2023Unifying Correspondence, Pose and NeRF for Pose-Free Novel View Synthesis from Stereo Pairscited by 37
  12. 2026HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worldscited by 34
  13. 2026LingBot-Map: Geometric Context Transformer for Streaming 3D Reconstructioncited by 32
  14. 2025DUNE: Distilling a Universal Encoder from Heterogeneous 2D and 3D Teacherscited by 28
  15. 2025FG2: Fine-Grained Cross-View Localization by Fine-Grained Feature Matchingcited by 27
  16. 2024Matching 2D Images in 3D: Metric Relative Pose from Metric Correspondencescited by 25
  17. 2026Scal3R: Scalable Test-Time Training for Large-Scale 3D Reconstructioncited by 25
  18. 2026ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Trainingcited by 23
  19. 2023Kick Back & Relax: Learning to Reconstruct the World by Watching SlowTVcited by 21
  20. 2024The Unreasonable Effectiveness of Pre-Trained Features for Camera Pose Refinementcited by 21
  21. 2026VGGT-Omegacited by 20
  22. 2025AMB3R: Accurate Feed-forward Metric-scale 3D Reconstruction with Backendcited by 16
  23. 2026WorldStereo: Bridging Camera-Guided Video Generation and Scene Reconstruction via 3D Geometric Memoriescited by 15
  24. 2023SACReg: Scene-Agnostic Coordinate Regression for Visual Localizationcited by 14
  25. 2026VGG-T3: Offline Feed-Forward 3D Reconstruction at Scalecited by 14
  26. 2024FAR: Flexible, Accurate and Robust 6DoF Relative Camera Pose Estimationcited by 12
  27. 2023Two-View Geometry Scoring Without Correspondencescited by 11
  28. 2025Rig3R: Rig-Aware Conditioning for Learned 3D Reconstructioncited by 11
  29. 2025G-CUT3R: Guided 3D Reconstruction with Camera and Depth Prior Integrationcited by 11
  30. 2025A Guide to Structureless Visual Localizationcited by 9
  31. 2023Lazy Visual Localization via Motion Averagingcited by 8
  32. 2025RUBIK: A Structured Benchmark for Image Matching across Geometric Challengescited by 8
  33. 2025ACE-G: Improving Generalization of Scene Coordinate Regression Through Query Pre-Trainingcited by 8
  34. 2026Flow4R: Unifying 4D Reconstruction and Tracking with Scene Flowcited by 8
  35. 2024Kick Back & Relax++: Scaling Beyond Ground-Truth Depth with SlowTV & CribsTVcited by 7
  36. 2025Deep Learning Reforms Image Matching: A Survey and Outlookcited by 7
  37. 2026LoMa: Local Feature Matching Revisitedcited by 7
  38. 2022Leveraging Image Matching Toward End-to-End Relative Camera Pose Regressioncited by 6

50 most cited of 112 All 112 (JSON)