mlsanityData to 7 Oct 2026

Toward Geometric Deep SLAM

arXiv 1707.07410 ↗First posted 24 Jul 2017cs.CVcited by 38

Not on the map. Why?

Abstract (arXiv)

We present a point tracking system powered by two deep convolutional neural networks. The first network, MagicPoint, operates on single images and extracts salient 2D points. The extracted points are "SLAM-ready" because they are by design isolated and well-distributed throughout the image. We compare this network against classical point detectors and discover a significant performance gap in the presence of image noise. As transformation estimation is more simple when the detected points are geometrically stable, we designed a second network, MagicWarp, which operates on pairs of point images (outputs of MagicPoint), and estimates the homography that relates the inputs. This transformation engine differs from traditional approaches because it does not use local point descriptors, only point locations. Both networks are trained with simple synthetic data, alleviating the requirement of expensive external camera ground truthing and advanced graphics rendering pipelines. The system is fast and lean, easily running 30+ FPS on a single CPU.

Key facts. Cited by 38 papers in the map (data through 7 Oct 2026). First posted on arXiv 24 Jul 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 9; 2022: 9; 2023: 8; 2024: 5; 2025: 6; 2026: 1.

38 papers.

Sort:

  1. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
  2. 2022OnePose: One-Shot Object Pose Estimation without CAD Modelscited by 155
  3. 2023Doppelgangers: Learning to Disambiguate Images of Similar Structurescited by 32
  4. 2022Self-Supervised Equivariant Learning for Oriented Keypoint Detectioncited by 26
  5. 2023Learning Rotation-Equivariant Features for Visual Correspondencecited by 23
  6. 2023Are Local Features All You Need for Cross-Domain Visual Place Recognition?cited by 21
  7. 2021DFM: A Performance Baseline for Deep Feature Matchingcited by 17
  8. 2021Advances in Inference and Representation for Simultaneous Localization and Mappingcited by 16
  9. 2021Overcoming Obstructions via Bandwidth-Limited Multi-Agent Spatial Handshakingcited by 16
  10. 2023Improving Transformer-based Image Matching by Cascaded Capturing Spatially Informative Keypointscited by 15
  11. 20223DG-STFM: 3D Geometric Guided Student-Teacher Feature Matchingcited by 14
  12. 2022Learning Robotic Navigation from Experience: Principles, Methods, and Recent Resultscited by 13
26 more
  1. 2021Graph-based Thermal-Inertial SLAM with Probabilistic Neural Networkscited by 12
  2. 2022GeoFill: Reference-Based Image Inpainting with Better Geometric Understandingcited by 9
  3. 2025Deep Learning Reforms Image Matching: A Survey and Outlookcited by 7
  4. 2022A Light Touch Approach to Teaching Transformers Multi-view Geometrycited by 6
  5. 2023Handbook on Leveraging Lines for Two-View Relative Pose Estimationcited by 6
  6. 2021Effect of Parameter Optimization on Classical and Learning-based Image Matching Methodscited by 5
  7. 2021Anomaly Detection of Defect using Energy of Point Pattern Features within Random Finite Set Frameworkcited by 5
  8. 2023AffineGlue: Joint Matching and Robust Estimationcited by 5
  9. 2023Occ2Net: Robust Image Matching Based on 3D Occupancy Estimation for Occluded Regionscited by 5
  10. 2024SRPose: Two-view Relative Pose Estimation with Sparse Keypointscited by 5
  11. 2024Sonar-based Deep Learning in Underwater Robotics: Overview, Robustness and Challengescited by 5
  12. 2025MapGlue: Multimodal Remote Sensing Image Matchingcited by 5
  13. 2022OpenGlue: Open Source Graph Neural Net Based Pipeline for Image Matchingcited by 3
  14. 20223DFill:Reference-guided Image Inpainting by Self-supervised 3D Image Alignmentcited by 3
  15. 2023Learning Probabilistic Coordinate Fields for Robust Correspondencescited by 3
  16. 2025Minimal Solvers for Full DoF Motion Estimation from Asynchronous Trackscited by 2
  17. 2021Learning and Crafting for the Wide Multiple Baseline Stereocited by 1
  18. 2022Learning-Based Keypoint Registration for Fetoscopic Mosaickingcited by 1
  19. 2024Generalized Correspondence Matching via Flexible Hierarchical Refinement and Patch Descriptor Distillationcited by 1
  20. 2024GIMS: Image Matching System Based on Adaptive Graph Construction and Graph Neural Networkcited by 1
  21. 2025Robust and Real-time Surface Normal Estimation from Stereo Disparities using Affine Transformationscited by 1
  22. 2021Enhancing Multi-Robot Perception via Learned Data Associationcited by 0
  23. 2024Learning Visual Information Utility with PIXERcited by 0
  24. 2025Correspondence-Free Pose Estimation with Patterns: A Unified Approach for Multi-Dimensional Visioncited by 0
  25. 2025Unlocking Zero-shot Potential of Semi-dense Image Matching via Gaussian Splattingcited by 0
  26. 2026SGFormer: Structure-Guided Transformer for Robust Local Feature Matchingcited by 0