SuperGlue: Learning Feature Matching with Graph Neural Networks

Paul-Edouard Sarlin, Daniel DeTone, Tomasz Malisiewicz, Andrew Rabinovich
11/26/2019
cs.CV

Official code: magicleap/SuperGluePretrainedNetwork (4,099 stars, last commit 2024-08-30).

Abstract

This paper introduces SuperGlue, a neural network that matches two sets of local features by jointly finding correspondences and rejecting non-matchable points. Assignments are estimated by solving a differentiable optimal transport problem, whose costs are predicted by a graph neural network. We introduce a flexible context aggregation mechanism based on attention, enabling SuperGlue to reason about the underlying 3D scene and feature assignments jointly. Compared to traditional, hand-designed heuristics, our technique learns priors over geometric transformations and regularities of the 3D world through end-to-end training from image pairs. SuperGlue outperforms other learned approaches and achieves state-of-the-art results on the task of pose estimation in challenging real-world indoor and outdoor environments. The proposed method performs matching in real-time on a modern GPU and can be readily integrated into modern SfM or SLAM systems. The code and trained weights are publicly available at https://github.com/magicleap/SuperGluePretrainedNetwork.

View on arXivView PDF

Code Implementations(1)

magicleap/SuperGluePretrainedNetworkOfficial100%

SuperGlue: Learning Feature Matching with Graph Neural Networks (CVPR 2020, Oral)

4,099760PythonMar 17, 20202 years agoNOASSERTION
deep-learningfeature-matchinggraph-neural-networkspose-estimation

Similar Papers

Str-L Pose: Integrating Point and Structured Line for Relative Pose Estimation in Dual-Graph

86% match2024

CGI-Stereo: Accurate and Real-Time Stereo Matching via Context and Geometry Interaction

85% match1 repo2882023

LightGlue: Local Feature Matching at Light Speed

85% match3 repos4.8k2023

DeepMatcher: A Deep Transformer-based Network for Robust and Accurate Local Feature Matching

85% match1 repo1472023

DynamicGlue: Epipolar and Time-Informed Data Association in Dynamic Environments using Graph Neural Networks

85% match2024

XFeat: Accelerated Features for Lightweight Image Matching

84% match2 repos12024

Cite this paper

@article{sarlin2019superglue,
  title  = {SuperGlue: Learning Feature Matching with Graph Neural Networks},
  author = {Paul-Edouard Sarlin and Daniel DeTone and Tomasz Malisiewicz and Andrew Rabinovich},
  year   = {2019},
  eprint = {1911.11763},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1911.11763}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1911.11763.svg)](https://codeofpaper.com/paper/1911.11763)

Discussion