mlsanityData to 7 Oct 2026

Learning Affine Correspondences by Integrating Geometric Constraints

arXiv 2504.04834 ↗First posted 7 Apr 2025cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

Affine correspondences have received significant attention due to their benefits in tasks like image matching and pose estimation. Existing methods for extracting affine correspondences still have many limitations in terms of performance; thus, exploring a new paradigm is crucial. In this paper, we present a new pipeline designed for extracting accurate affine correspondences by integrating dense matching and geometric constraints. Specifically, a novel extraction framework is introduced, with the aid of dense matching and a novel keypoint scale and orientation estimator. For this purpose, we propose loss functions based on geometric constraints, which can effectively improve accuracy by supervising neural networks to learn feature geometry. The experimental show that the accuracy and robustness of our method outperform the existing ones in image matching tasks. To further demonstrate the effectiveness of the proposed method, we applied it to relative pose estimation. Affine correspondences extracted by our method lead to more accurate poses than the baselines on a range of real-world datasets. The code is available at https://github.com/stilcrad/DenseAffine.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 35 of its 65 references that are in the map. First posted on arXiv 7 Apr 2025.

Builds on

35 of 65 references.

References · 35

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  3. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  4. 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,313
  5. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
  6. 2018MegaDepth: Learning Single-View Depth Prediction from Internet Photoscited by 731
  7. 2019D2-Net: A Trainable CNN for Joint Detection and Description of Local Featurescited by 324
  8. 2020DISK: Learning local features with policy gradientcited by 310
  9. 2023RoMa: Robust Dense Feature Matchingcited by 296
  10. 2017HPatches: A benchmark and evaluation of handcrafted and learned local descriptorscited by 261
  11. 2016Large-Scale Image Retrieval with Attentive Deep Local Featurescited by 238
  12. 2020Image Matching across Wide Baselines: From Paper to Practicecited by 224
23 more
  1. 2017Working hard to know your neighbor's margins: Local descriptor learning losscited by 212
  2. 2022DKM: Dense Kernelized Feature Matching for Geometry Estimationcited by 174
  3. 2022ASpanFormer: Detector-Free Image Matching with Adaptive Span Transformercited by 170
  4. 2019SOSNet: Second Order Similarity Regularization for Local Descriptor Learningcited by 131
  5. 2020ASLFeat: Learning Local Features of Accurate Shape and Localizationcited by 129
  6. 2020Patch2Pix: Epipolar-Guided Pixel-Level Correspondencescited by 128
  7. 2024Efficient LoFTR: Semi-Dense Local Feature Matching with Sparse-Like Speedcited by 119
  8. 2019Key.Net: Keypoint Detection by Handcrafted and Learned CNN Filterscited by 117
  9. 2020Learning Feature Descriptors using Camera Pose Supervisioncited by 104
  10. 2021PDC-Net+: Enhanced Probabilistic Dense Correspondence Networkcited by 80
  11. 2019ContextDesc: Local Descriptor Augmentation with Cross-Modality Contextcited by 79
  12. 2024GIM: Learning Generalizable Image Matcher From Internet Videoscited by 72
  13. 2023DeDoDe: Detect, Don't Describe -- Describe, Don't Detect for Local Feature Matchingcited by 62
  14. 2017Repeatability Is Not Enough: Learning Affine Regions via Discriminabilitycited by 57
  15. 2018GeoDesc: Learning Local Descriptors by Integrating Geometry Constraintscited by 52
  16. 2022Decoupling Makes Weakly Supervised Local Feature Bettercited by 28
  17. 2015MODS: Fast and Robust Method for Two-View Matchingcited by 26
  18. 2024Local Feature Matching Using Deep Learning: A Surveycited by 26
  19. 2017A Minimal Solution for Two-view Focal-length Estimation using Two Affine Correspondencescited by 14
  20. 2020Making Affine Correspondences Work in Camera Geometry Computationcited by 13
  21. 2023Affine Correspondences between Multi-Camera Systems for Relative Pose Estimationcited by 13
  22. 2020Minimal Cases for Computing the Generalized Relative Pose using Affine Correspondencescited by 7
  23. 2019Optimal Multi-view Correction of Local Affine Framescited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 2.

5 papers.

  1. 2025Deep Learning Reforms Image Matching: A Survey and Outlookcited by 7
  2. 2025DistillMatch: Leveraging Knowledge Distillation from Vision Foundation Model for Multimodal Image Matchingcited by 4
  3. 2025A Minimal Solver for Relative Pose Estimation with Unknown Focal Length from Two Affine Correspondencescited by 1
  4. 2026MatRes: Zero-Shot Test-Time Model Adaptation for Simultaneous Matching and Restorationcited by 0
  5. 2026Rolling Shutter Relative Pose Estimation Made Practicalcited by 0