magicleap/SuperGluePretrainedNetwork

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

51
/ 100
Established

Combines Graph Neural Networks with an Optimal Matching layer to perform context-aware feature matching across image pairs, operating as an end-to-end "middle-end" that aggregates spatial context before matching. Provides pretrained indoor (ScanNet) and outdoor (MegaDepth) models that work on top of SuperPoint keypoints, with PyTorch inference code and evaluation utilities including RANSAC-based pose estimation and matching confidence scoring. Integrates with the hloc visual localization toolbox and supports live webcam/video streams, image sequences, and batch pair processing for both qualitative visualization and quantitative evaluation against ground truth poses.

3,967 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 25 / 25

How are scores calculated?

Stars

3,967

Forks

762

Language

Python

License

Last pushed

Aug 30, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/magicleap/SuperGluePretrainedNetwork"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.