mlsanityData to 7 Oct 2026

LoFTR: Detector-Free Local Feature Matching with Transformers

arXiv 2104.00680 ↗First posted 1 Apr 2021cs.CV cs.ROcited by 974

Not on the map. Why?

Cited by 28 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We present a novel method for local image feature matching. Instead of performing image feature detection, description, and matching sequentially, we propose to first establish pixel-wise dense matches at a coarse level and later refine the good matches at a fine level. In contrast to dense methods that use a cost volume to search correspondences, we use self and cross attention layers in Transformer to obtain feature descriptors that are conditioned on both images. The global receptive field provided by Transformer enables our method to produce dense matches in low-texture areas, where feature detectors usually struggle to produce repeatable interest points. The experiments on indoor and outdoor datasets show that LoFTR outperforms state-of-the-art methods by a large margin. LoFTR also ranks first on two public benchmarks of visual localization among the published methods.

Key facts. Cited by 974 papers in the map (data through 7 Oct 2026). Builds on 40 of its 58 references that are in the map. First posted on arXiv 1 Apr 2021.

Builds on

40 of 58 references.

References · 40

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2020End-to-End Object Detection with Transformerscited by 7,384
  5. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  6. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  7. 2020Transformers are RNNs: Fast Autoregressive Transformers with Linear Attentioncited by 1,979
  8. 2020Reformer: The Efficient Transformercited by 1,658
  9. 2020Rethinking Attention with Performerscited by 1,584
  10. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  11. 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,313
  12. 2020Efficient Transformers: A Surveycited by 784
28 more
  1. 2018MegaDepth: Learning Single-View Depth Prediction from Internet Photoscited by 731
  2. 2018From Coarse to Fine: Robust Hierarchical Localization at Large Scalecited by 556
  3. 2020Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentationcited by 326
  4. 2020DISK: Learning local features with policy gradientcited by 310
  5. 2016LIFT: Learned Invariant Feature Transformcited by 305
  6. 2017HPatches: A benchmark and evaluation of handcrafted and learned local descriptorscited by 261
  7. 2019R2D2: Repeatable and Reliable Detector and Descriptorcited by 206
  8. 2020Revisiting Stereo Depth Estimation From a Sequence-to-Sequence Perspective with Transformerscited by 190
  9. 2021COTR: Correspondence Transformer for Matching Across Imagescited by 182
  10. 2017Learning to Find Good Correspondencescited by 176
  11. 2018Neighbourhood Consensus Networkscited by 173
  12. 2019Learning Two-View Correspondences and Geometry Using Order-Aware Networkcited by 148
  13. 2019GLU-Net: Global-Local Universal Network for Dense Flow and Correspondencescited by 139
  14. 2020ASLFeat: Learning Local Features of Accurate Shape and Localizationcited by 129
  15. 2021Learning Accurate Dense Correspondences and When to Trust Themcited by 113
  16. 2020Learning Feature Descriptors using Camera Pose Supervisioncited by 104
  17. 2020Efficient Neighbourhood Consensus Networks via Submanifold Sparse Convolutionscited by 92
  18. 2020Dual-Resolution Correspondence Networkscited by 90
  19. 2016Universal Correspondence Networkcited by 82
  20. 2019ContextDesc: Local Descriptor Augmentation with Cross-Modality Contextcited by 79
  21. 2020Robust Image Retrieval-based Visual Localization using Kapturecited by 66
  22. 2020GOCor: Bringing Globally Optimized Correspondence Volumes into Your Neural Networkcited by 53
  23. 2019GIFT: Learning Transformation-Invariant Dense Visual Descriptors via Group CNNscited by 47
  24. 2020Online Invariance Selection for Local Feature Descriptorscited by 41
  25. 2017Toward Geometric Deep SLAMcited by 38
  26. 2020Extremely Dense Point Correspondences using a Learned Feature Descriptorcited by 20
  27. 2020Image Stylization for Robust Featurescited by 11
  28. 2018InLoc: Indoor Visual Localization with Dense Matching and View Synthesiscited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 28; 2022: 105; 2023: 192; 2024: 198; 2025: 253; 2026: 198.

974 papers.

Sort:

By area:

  1. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,526
  2. 2025VGGT: Visual Geometry Grounded Transformercited by 1,475
  3. 2024Grounding Image Matching in 3D with MASt3Rcited by 960
  4. 2023LightGlue: Local Feature Matching at Light Speedcited by 583
  5. 2023One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape Optimizationcited by 555
  6. 2024BLINK: Multimodal Large Language Models Can See but Not Perceivecited by 529
  7. 2025Continuous 3D Perception Model with Persistent Statecited by 527
  8. 2022TransFusion: Robust LiDAR-Camera Fusion for 3D Object Detection with Transformerscited by 464
  9. 2023FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objectscited by 411
  10. 2021Deep ViT Features as Dense Visual Descriptorscited by 327
  11. 2021GMFlow: Learning Optical Flow via Global Matchingcited by 324
  12. 2023Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Drivingcited by 309
38 more
  1. 2023RoMa: Robust Dense Feature Matchingcited by 296
  2. 2022FlowFormer: A Transformer Architecture for Optical Flowcited by 273
  3. 20243D Reconstruction with Spatial Memorycited by 269
  4. 2023Visual Geometry Grounded Deep Structure From Motioncited by 225
  5. 2024Motion-I2V: Consistent and Controllable Image-to-Video Generation with Explicit Motion Modelingcited by 209
  6. 2022Geometric Transformer for Fast and Robust Point Cloud Registrationcited by 200
  7. 2021Pixel-Perfect Structure-from-Motion with Featuremetric Refinementcited by 199
  8. 2022Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlationcited by 197
  9. 2023Tracking Everything Everywhere All at Oncecited by 190
  10. 2022Deep Patch Visual Odometrycited by 186
  11. 2025FLARE: Feed-forward Geometry, Appearance and Camera Estimation from Uncalibrated Sparse Viewscited by 176
  12. 2024Probing the 3D Awareness of Visual Foundation Modelscited by 175
  13. 2022DKM: Dense Kernelized Feature Matching for Geometry Estimationcited by 174
  14. 2022ASpanFormer: Detector-Free Image Matching with Adaptive Span Transformercited by 170
  15. 2024SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flowcited by 169
  16. 2023Panacea: Panoramic and Controllable Video Generation for Autonomous Drivingcited by 165
  17. 2024MASt3R-SfM: a Fully-Integrated Solution for Unconstrained Structure-from-Motioncited by 161
  18. 2022OnePose: One-Shot Object Pose Estimation without CAD Modelscited by 155
  19. 2021CoFiNet: Reliable Coarse-to-fine Correspondences for Robust Point Cloud Registrationcited by 151
  20. 2023Optimal Transport Aggregation for Visual Place Recognitioncited by 151
  21. 2023DynamicStereo: Consistent Dynamic Depth from Stereo Videoscited by 147
  22. 2023BundleSDF: Neural 6-DoF Tracking and 3D Reconstruction of Unknown Objectscited by 144
  23. 2022QuadTree Attention for Vision Transformerscited by 129
  24. 2022MatchFormer: Interleaving Attention in Transformers for Feature Matchingcited by 126
  25. 2023ALIKED: A Lighter Keypoint and Descriptor Extraction Network via Deformable Transformationcited by 120
  26. 2023Unsupervised Semantic Correspondence Using Stable Diffusioncited by 120
  27. 2024Efficient LoFTR: Semi-Dense Local Feature Matching with Sparse-Like Speedcited by 119
  28. 2021TransMVSNet: Global Context-aware Multi-view Stereo Network with Transformerscited by 117
  29. 2023SAM-6D: Segment Anything Model Meets Zero-Shot 6D Object Pose Estimationcited by 115
  30. 2023OnePose++: Keypoint-Free One-Shot Object Pose Estimation without CAD Modelscited by 114
  31. 2024XFeat: Accelerated Features for Lightweight Image Matchingcited by 113
  32. 2022Map-free Visual Relocalization: Metric Pose Relative to a Single Imagecited by 112
  33. 2023EigenPlaces: Training Viewpoint Robust Models for Visual Place Recognitioncited by 109
  34. 2023FoundPose: Unseen Object Pose Estimation with Foundation Featurescited by 109
  35. 2023Street-View Image Generation from a Bird's-Eye View Layoutcited by 102
  36. 2022LaMAR: Benchmarking Localization and Mapping for Augmented Realitycited by 98
  37. 2023WoVoGen: World Volume-aware Diffusion for Controllable Multi-camera Driving Scene Generationcited by 93
  38. 2022ParticleSfM: Exploiting Dense Point Trajectories for Localizing Moving Cameras in the Wildcited by 90

50 most cited of 974 All 974 (JSON)