gbarbarani/re-ranking-for-VPR

Code for "Are Local Features All You Need for Cross-Domain Visual Place Recognition?" CVPR IMW 2023.

23
/ 100
Experimental

This project helps self-driving cars and drone navigation systems identify their location accurately by comparing a live camera feed (query image) with a database of known locations. It takes in collections of images, each tagged with its precise geographic coordinates, and outputs a refined list of potential matching locations, improving the system's ability to recognize where it is, even under challenging conditions like night-time or different seasons. Robotics engineers and autonomous system developers would find this useful for improving navigation reliability.

No commits in the last 6 months.

Use this if you need to significantly improve the accuracy of visual place recognition for autonomous systems operating in diverse, real-world environments with varying conditions.

Not ideal if you are looking for a simple, off-the-shelf navigation solution without the need for advanced re-ranking or cross-domain performance improvements.

autonomous-navigation robotics-localization drone-mapping visual-place-recognition geographic-information-systems
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 7 / 25

How are scores calculated?

Stars

60

Forks

3

Language

Python

License

Last pushed

Aug 03, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/gbarbarani/re-ranking-for-VPR"

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