AnyLoc/Revisit-Anything

Code release for Revisit Anything: Visual Place Recognition via Image Segment Retrieval (ECCV 2024)

40
/ 100
Emerging

This project helps self-driving cars, drone navigation systems, and augmented reality applications recognize specific locations by analyzing images. It takes raw image data, breaks it down into individual objects and regions, and then identifies if that location has been seen before. The output tells you if a new image matches a known place, even if the viewpoint has changed or the scene is crowded. It's designed for engineers developing navigation and localization systems.

166 stars. No commits in the last 6 months.

Use this if you need highly accurate visual place recognition for autonomous systems, especially in diverse environments like urban streets, indoor malls, or even historical aerial views.

Not ideal if your primary goal is general object detection or scene classification rather than precise location matching from varying perspectives.

autonomous-navigation robotics visual-localization SLAM geospatial-mapping
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

166

Forks

14

Language

Jupyter Notebook

License

BSD-3-Clause

Last pushed

Aug 02, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/AnyLoc/Revisit-Anything"

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