#217 · Primary category： Computer Vision
SuperGluePretrainedNetwork
SuperGlue: Learning Feature Matching with Graph Neural Networks (CVPR 2020, Oral)
Project last updated：08/30/24
GitHub Stars
4.1K
Forks
758
Contributors
9
License
Other
Why we included this project
SuperGlue is the reference implementation of Magic Leap's CVPR 2020 paper, a learned middle-end that takes sparse keypoints and descriptors (SuperPoint works well) and produces clean matches between image pairs. That makes it a practical tool for visual localization, structure-from-motion, image stitching, and pose estimation, where cutting wrong correspondences saves hours of manual tuning. The repo includes PyTorch code, pretrained indoor and outdoor weights, a live webcam demo, and a batch script that writes matches to disk and can evaluate them against ground-truth relative poses with RANSAC essential-matrix decomposition. It's also a readable baseline worth studying or adapting, though the license restricts it to noncommercial academic research.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
opencv
Open Source Computer Vision Library
RuView
π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video.
PaddleOCR
Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.
MinerU
Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
tesseract
Tesseract Open Source OCR Engine (main repository)