mlsanityData to 7 Oct 2026

D2-Net: A Trainable CNN for Joint Detection and Description of Local Features

arXiv 1905.03561 ↗First posted 9 May 2019cs.CVcited by 324

Not on the map. Why?

Abstract (arXiv)

In this work we address the problem of finding reliable pixel-level correspondences under difficult imaging conditions. We propose an approach where a single convolutional neural network plays a dual role: It is simultaneously a dense feature descriptor and a feature detector. By postponing the detection to a later stage, the obtained keypoints are more stable than their traditional counterparts based on early detection of low-level structures. We show that this model can be trained using pixel correspondences extracted from readily available large-scale SfM reconstructions, without any further annotations. The proposed method obtains state-of-the-art performance on both the difficult Aachen Day-Night localization dataset and the InLoc indoor localization benchmark, as well as competitive performance on other benchmarks for image matching and 3D reconstruction.

Key facts. Cited by 324 papers in the map (data through 7 Oct 2026). First posted on arXiv 9 May 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 41; 2022: 56; 2023: 68; 2024: 53; 2025: 69; 2026: 37.

324 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. 2025Depth Anything 3: Recovering the Visual Space from Any Viewscited by 626
  5. 2025Continuous 3D Perception Model with Persistent Statecited by 527
  6. 2025Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Passcited by 356
  7. 2023A Tale of Two Features: Stable Diffusion Complements DINO for Zero-Shot Semantic Correspondencecited by 297
  8. 2021Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognitioncited by 236
  9. 2023Visual Geometry Grounded Deep Structure From Motioncited by 225
  10. 2023Tracking Everything Everywhere All at Oncecited by 190
  11. 2021COTR: Correspondence Transformer for Matching Across Imagescited by 182
  12. 2025FLARE: Feed-forward Geometry, Appearance and Camera Estimation from Uncalibrated Sparse Viewscited by 176
38 more
  1. 2022ASpanFormer: Detector-Free Image Matching with Adaptive Span Transformercited by 170
  2. 2023PF-LRM: Pose-Free Large Reconstruction Model for Joint Pose and Shape Predictioncited by 165
  3. 2021Learning Target Candidate Association to Keep Track of What Not to Trackcited by 131
  4. 2023ALIKED: A Lighter Keypoint and Descriptor Extraction Network via Deformable Transformationcited by 120
  5. 2024Efficient LoFTR: Semi-Dense Local Feature Matching with Sparse-Like Speedcited by 119
  6. 2021Learning Multi-Scene Absolute Pose Regression with Transformerscited by 108
  7. 2022TransVPR: Transformer-based place recognition with multi-level attention aggregationcited by 102
  8. 2021Learning to Match Features with Seeded Graph Matching Networkcited by 92
  9. 2022ParticleSfM: Exploiting Dense Point Trajectories for Localizing Moving Cameras in the Wildcited by 90
  10. 2025Point3R: Streaming 3D Reconstruction with Explicit Spatial Pointer Memorycited by 82
  11. 2024Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localizationcited by 81
  12. 2021ALIKE: Accurate and Lightweight Keypoint Detection and Descriptor Extractioncited by 77
  13. 2024Scene Coordinate Reconstruction: Posing of Image Collections via Incremental Learning of a Relocalizercited by 73
  14. 2024GIM: Learning Generalizable Image Matcher From Internet Videoscited by 72
  15. 2021Deep Learning for Instance Retrieval: A Surveycited by 70
  16. 2021Robust Reference-based Super-Resolution via C2-Matchingcited by 68
  17. 2022CATs++: Boosting Cost Aggregation with Convolutions and Transformerscited by 53
  18. 2022EndoMapper dataset of complete calibrated endoscopy procedurescited by 53
  19. 2024XoFTR: Cross-modal Feature Matching Transformercited by 45
  20. 2025LONG3R: Long Sequence Streaming 3D Reconstructioncited by 45
  21. 2021P2-Net: Joint Description and Detection of Local Features for Pixel and Point Matchingcited by 42
  22. 2021Warp Consistency for Unsupervised Learning of Dense Correspondencescited by 42
  23. 2022IFOR: Iterative Flow Minimization for Robotic Object Rearrangementcited by 42
  24. 2024FreeSplatter: Pose-free Gaussian Splatting for Sparse-view 3D Reconstructioncited by 40
  25. 2025MatchAnything: Universal Cross-Modality Image Matching with Large-Scale Pre-Trainingcited by 40
  26. 2025Matrix3D: Large Photogrammetry Model All-in-Onecited by 39
  27. 2021Deep Lucas-Kanade Homography for Multimodal Image Alignmentcited by 38
  28. 2022Visual Localization via Few-Shot Scene Region Classificationcited by 36
  29. 2023Tracking and Mapping in Medical Computer Vision: A Reviewcited by 36
  30. 2023Differentiable Registration of Images and LiDAR Point Clouds with VoxelPoint-to-Pixel Matchingcited by 35
  31. 2022BALF: Simple and Efficient Blur Aware Local Feature Detectorcited by 34
  32. 2023PATS: Patch Area Transportation with Subdivision for Local Feature Matchingcited by 33
  33. 2021Hyperdimensional computing as a framework for systematic aggregation of image descriptorscited by 32
  34. 2022TopicFM: Robust and Interpretable Topic-Assisted Feature Matchingcited by 32
  35. 2024The NeRFect Match: Exploring NeRF Features for Visual Localizationcited by 32
  36. 2023Visual Place Recognition: A Tutorialcited by 31
  37. 2024GS-CPR: Efficient Camera Pose Refinement via 3D Gaussian Splattingcited by 31
  38. 2022Adaptive Assignment for Geometry Aware Local Feature Matchingcited by 30

50 most cited of 324 All 324 (JSON)