mlsanityData to 7 Oct 2026

RUBIK: A Structured Benchmark for Image Matching across Geometric Challenges

arXiv 2502.19955 ↗First posted 27 Feb 2025cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

Camera pose estimation is crucial for many computer vision applications, yet existing benchmarks offer limited insight into method limitations across different geometric challenges. We introduce RUBIK, a novel benchmark that systematically evaluates image matching methods across well-defined geometric difficulty levels. Using three complementary criteria - overlap, scale ratio, and viewpoint angle - we organize 16.5K image pairs from nuScenes into 33 difficulty levels. Our comprehensive evaluation of 14 methods reveals that while recent detector-free approaches achieve the best performance (>47% success rate), they come with significant computational overhead compared to detector-based methods (150-600ms vs. 40-70ms). Even the best performing method succeeds on only 54.8% of the pairs, highlighting substantial room for improvement, particularly in challenging scenarios combining low overlap, large scale differences, and extreme viewpoint changes. Benchmark will be made publicly available.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 40 of its 60 references that are in the map. First posted on arXiv 27 Feb 2025.

Builds on

40 of 60 references.

References · 40

  1. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  2. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  3. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,526
  4. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  5. 2024Depth Anything V2cited by 1,385
  6. 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,313
  7. 2015ORB-SLAM: a Versatile and Accurate Monocular SLAM Systemcited by 983
  8. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
  9. 2024Grounding Image Matching in 3D with MASt3Rcited by 960
  10. 2016ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Camerascited by 801
  11. 2018MegaDepth: Learning Single-View Depth Prediction from Internet Photoscited by 731
  12. 2023LightGlue: Local Feature Matching at Light Speedcited by 583
28 more
  1. 2016Past, Present, and Future of Simultaneous Localization And Mapping: Towards the Robust-Perception Agecited by 405
  2. 2024UniDepth: Universal Monocular Metric Depth Estimationcited by 375
  3. 2020DISK: Learning local features with policy gradientcited by 310
  4. 2023RoMa: Robust Dense Feature Matchingcited by 296
  5. 2017HPatches: A benchmark and evaluation of handcrafted and learned local descriptorscited by 261
  6. 2020Image Matching across Wide Baselines: From Paper to Practicecited by 224
  7. 2021COTR: Correspondence Transformer for Matching Across Imagescited by 182
  8. 2022DKM: Dense Kernelized Feature Matching for Geometry Estimationcited by 174
  9. 2022ASpanFormer: Detector-Free Image Matching with Adaptive Span Transformercited by 170
  10. 2019MAGSAC++, a fast, reliable and accurate robust estimatorcited by 160
  11. 2022QuadTree Attention for Vision Transformerscited by 129
  12. 2022MatchFormer: Interleaving Attention in Transformers for Feature Matchingcited by 126
  13. 2023ALIKED: A Lighter Keypoint and Descriptor Extraction Network via Deformable Transformationcited by 120
  14. 2024Efficient LoFTR: Semi-Dense Local Feature Matching with Sparse-Like Speedcited by 119
  15. 2024XFeat: Accelerated Features for Lightweight Image Matchingcited by 113
  16. 2022Map-free Visual Relocalization: Metric Pose Relative to a Single Imagecited by 112
  17. 2021Learning to Match Features with Seeded Graph Matching Networkcited by 92
  18. 2023SiLK -- Simple Learned Keypointscited by 54
  19. 2024DeDoDe v2: Analyzing and Improving the DeDoDe Keypoint Detectorcited by 38
  20. 2022TopicFM: Robust and Interpretable Topic-Assisted Feature Matchingcited by 32
  21. 2020Deep Keypoint-Based Camera Pose Estimation with Geometric Constraintscited by 29
  22. 2022ECO-TR: Efficient Correspondences Finding Via Coarse-to-Fine Refinementcited by 18
  23. 2021Neural Reprojection Error: Merging Feature Learning and Camera Pose Estimationcited by 17
  24. 2023Unsupervised Keypoints from Pretrained Diffusion Modelscited by 15
  25. 20223DG-STFM: 3D Geometric Guided Student-Teacher Feature Matchingcited by 14
  26. 2024Learning to Make Keypoints Sub-Pixel Accuratecited by 13
  27. 2023Occ2Net: Robust Image Matching Based on 3D Occupancy Estimation for Occluded Regionscited by 5
  28. 2024SRPose: Two-view Relative Pose Estimation with Sparse Keypointscited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 5.

8 papers.

  1. 2025RoMa v2: Harder Better Faster Denser Feature Matchingcited by 44
  2. 2025Alligat0R: Pre-Training Through Co-Visibility Segmentation for Relative Camera Pose Regressioncited by 10
  3. 2025Deep Learning Reforms Image Matching: A Survey and Outlookcited by 7
  4. 2026LoMa: Local Feature Matching Revisitedcited by 7
  5. 2026AirZoo: A Unified Large-Scale Dataset for Grounding Aerial Geometric 3D Visioncited by 3
  6. 2026Understanding and Optimizing Attention-Based Sparse Matching for Diverse Local Featurescited by 0
  7. 2026XFeat Revisited: Reproducibility and Evaluation of a Lightweight Image Matchercited by 0
  8. 2026RoMa-Omega: What Feed-Forward 3D Models Know About Image Matchingcited by 0