Scale-Free Image Keypoints Using Differentiable Persistent Homology
Not on the map. Why?
Abstract (arXiv)
In computer vision, keypoint detection is a fundamental task, with applications spanning from robotics to image retrieval; however, existing learning-based methods suffer from scale dependency and lack flexibility. This paper introduces a novel approach that leverages Morse theory and persistent homology, powerful tools rooted in algebraic topology. We propose a novel loss function based on the recent introduction of a notion of subgradient in persistent homology, paving the way toward topological learning. Our detector, MorseDet, is the first topology-based learning model for feature detection, which achieves competitive performance in keypoint repeatability and introduces a principled and theoretically robust approach to the problem.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 28 of its 60 references that are in the map. First posted on arXiv 3 Jun 2024.
Builds on
28 of 60 references.
References · 28
- 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,320
- 2015ORB-SLAM: a Versatile and Accurate Monocular SLAM Systemcited by 988
- 2018MegaDepth: Learning Single-View Depth Prediction from Internet Photoscited by 737
- 2018From Coarse to Fine: Robust Hierarchical Localization at Large Scalecited by 559
- 2020DISK: Learning local features with policy gradientcited by 312
- 2016LIFT: Learned Invariant Feature Transformcited by 306
- 2017HPatches: A benchmark and evaluation of handcrafted and learned local descriptorscited by 264
- 2016Large-Scale Image Retrieval with Attentive Deep Local Featurescited by 238
- 2019A Topological Loss Function for Deep-Learning based Image Segmentation using Persistent Homologycited by 134
- 2019Topological Autoencoderscited by 127
- 2023ALIKED: A Lighter Keypoint and Descriptor Extraction Network via Deformable Transformationcited by 121
- 2019Key.Net: Keypoint Detection by Handcrafted and Learned CNN Filterscited by 118
16 more
- 2017Deep Learning with Topological Signaturescited by 114
- 2021ALIKE: Accurate and Lightweight Keypoint Detection and Descriptor Extractioncited by 77
- 2021Topology-Aware Segmentation Using Discrete Morse Theorycited by 75
- 2018A Topological Regularizer for Classifiers via Persistent Homologycited by 63
- 2017Repeatability Is Not Enough: Learning Affine Regions via Discriminabilitycited by 57
- 2016Quad-networks: unsupervised learning to rank for interest point detectioncited by 36
- 2023Topology-Aware Uncertainty for Image Segmentationcited by 32
- 2019A Framework for Differential Calculus on Persistence Barcodescited by 29
- 2023Are Local Features All You Need for Cross-Domain Visual Place Recognition?cited by 21
- 2022Learning Probabilistic Topological Representations Using Discrete Morse Theorycited by 20
- 2017Persistent homology machine learning for fingerprint classificationcited by 11
- 2017Recent Advances in Features Extraction and Description Algorithms: A Comprehensive Surveycited by 8
- 2020FFD: Fast Feature Detectorcited by 5
- 2018Large scale evaluation of local image feature detectors on homography datasetscited by 4
- 2014Comparing Feature Detectors: A bias in the repeatability criteria, and how to correct itcited by 2
- 2018From handcrafted to deep local featurescited by 2
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.