DagnyT/hardnet

Hardnet descriptor model - "Working hard to know your neighbor's margins: Local descriptor learning loss"

52
/ 100
Established

This is a machine learning model that helps computer vision engineers and researchers extract robust features from images, making it easier to match distinct but similar image patches. It takes image patches as input and generates a numerical "descriptor" for each, which can then be used for tasks like image retrieval or object recognition. It's intended for those working on improving the accuracy of visual search or feature matching in their applications.

530 stars. No commits in the last 6 months.

Use this if you are a computer vision engineer or researcher focused on extracting highly discriminative local features from image patches to improve image matching and retrieval tasks.

Not ideal if you need a complete, end-to-end image retrieval system out-of-the-box, as this provides a core component (the descriptor) rather than a full solution.

computer-vision image-retrieval feature-matching object-recognition pattern-recognition
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 24 / 25

How are scores calculated?

Stars

530

Forks

102

Language

Python

License

MIT

Last pushed

Apr 14, 2025

Commits (30d)

0

Get this data via API

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

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