mlsanityData to 7 Oct 2026

SDGMNet: Statistic-based Dynamic Gradient Modulation for Local Descriptor Learning

arXiv 2106.04434 ↗First posted 8 Jun 2021cs.CV cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

Modifications on triplet loss that rescale the back-propagated gradients of special pairs have made significant progress on local descriptor learning. However, current gradient modulation strategies are mainly static so that they would suffer from changes of training phases or datasets. In this paper, we propose a dynamic gradient modulation, named SDGMNet, to improve triplet loss for local descriptor learning. The core of our method is formulating modulation functions with statistical characteristics which are estimated dynamically. Firstly, we perform deep analysis on back propagation of general triplet-based loss and introduce included angle for distance measure. On this basis, auto-focus modulation is employed to moderate the impact of statistically uncommon individual pairs in stochastic gradient descent optimization; probabilistic margin cuts off the gradients of proportional Siamese pairs that are believed to reach the optimum; power adjustment balances the total weights of negative pairs and positive pairs. Extensive experiments demonstrate that our novel descriptor surpasses previous state-of-the-arts on standard benchmarks including patch verification, matching and retrieval tasks.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 40 of its 81 references that are in the map. First posted on arXiv 8 Jun 2021.

Builds on

40 of 81 references.

References · 40

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  3. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  4. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  5. 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,313
  6. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
  7. 2018CosFace: Large Margin Cosine Loss for Deep Face Recognitioncited by 968
  8. 2017SphereFace: Deep Hypersphere Embedding for Face Recognitioncited by 733
  9. 2018MegaDepth: Learning Single-View Depth Prediction from Internet Photoscited by 731
  10. 2019Large-Scale Long-Tailed Recognition in an Open Worldcited by 704
  11. 2015Cyclical Learning Rates for Training Neural Networkscited by 627
  12. 2016MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognitioncited by 621
28 more
  1. 2016Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networkscited by 541
  2. 2015Deep Metric Learning via Lifted Structured Feature Embeddingcited by 449
  3. 2020Circle Loss: A Unified Perspective of Pair Similarity Optimizationcited by 402
  4. 2019Multi-Similarity Loss with General Pair Weighting for Deep Metric Learningcited by 368
  5. 2020CurricularFace: Adaptive Curriculum Learning Loss for Deep Face Recognitioncited by 320
  6. 2020A Metric Learning Reality Checkcited by 317
  7. 2020DISK: Learning local features with policy gradientcited by 310
  8. 2018Recent Advances in Open Set Recognition: A Surveycited by 309
  9. 2016LIFT: Learned Invariant Feature Transformcited by 305
  10. 2017HPatches: A benchmark and evaluation of handcrafted and learned local descriptorscited by 261
  11. 2017No Fuss Distance Metric Learning using Proxiescited by 225
  12. 2017Working hard to know your neighbor's margins: Local descriptor learning losscited by 212
  13. 2019R2D2: Repeatable and Reliable Detector and Descriptorcited by 206
  14. 2019SOSNet: Second Order Similarity Regularization for Local Descriptor Learningcited by 131
  15. 2020ASLFeat: Learning Local Features of Accurate Shape and Localizationcited by 129
  16. 2019Key.Net: Keypoint Detection by Handcrafted and Learned CNN Filterscited by 117
  17. 2017Deep Metric Learning with Angular Losscited by 107
  18. 2017L2-constrained Softmax Loss for Discriminative Face Verificationcited by 96
  19. 2018ArcFace: Additive Angular Margin Loss for Deep Face Recognitioncited by 56
  20. 2018GeoDesc: Learning Local Descriptors by Integrating Geometry Constraintscited by 52
  21. 2019Filter Response Normalization Layer: Eliminating Batch Dependence in the Training of Deep Neural Networkscited by 45
  22. 2020SOLAR: Second-Order Loss and Attention for Image Retrievalcited by 45
  23. 2018Local Descriptors Optimized for Average Precisioncited by 43
  24. 2019Beyond Cartesian Representations for Local Descriptorscited by 35
  25. 2017Learning Spread-out Local Feature Descriptorscited by 34
  26. 2020D2D: Keypoint Extraction with Describe to Detect Approachcited by 34
  27. 2015Learning Local Image Descriptors with Deep Siamese and Triplet Convolutional Networks by Minimising Global Loss Functionscited by 33
  28. 2017A Performance Evaluation of Local Features for Image Based 3D Reconstructioncited by 5

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 0; 2024: 0; 2025: 4.

5 papers.

  1. 2022ReDFeat: Recoupling Detection and Description for Multimodal Feature Learningcited by 26
  2. 2025Deep Learning Reforms Image Matching: A Survey and Outlookcited by 7
  3. 2025OSDM-MReg: Multimodal Image Registration based One Step Diffusion Modelcited by 2
  4. 2025Pore-scale Image Patch Dataset and A Comparative Evaluation of Pore-scale Facial Featurescited by 1
  5. 2025Grid-Reg: Detector-Free Gridized Feature Learning and Matching for Large-Scale SAR-Optical Image Registrationcited by 0