csiro-robotics/HOTFormerLoc

[IEEE/CVF CVPR 2025] Hierarchical Octree Transformer for Versatile Lidar Place Recognition Across Ground and Aerial Views

43
/ 100
Emerging

This project offers a robust solution for determining the precise location of a robot or autonomous vehicle by analyzing 3D lidar scans, even when comparing data from ground-level and aerial perspectives. It takes in 3D lidar point clouds from either ground or aerial sources and outputs highly accurate place recognition, allowing for reliable navigation and mapping in complex environments like forests or urban areas. This is ideal for robotics engineers, autonomous vehicle developers, or mapping specialists working with diverse lidar data.

Use this if you need to accurately identify a previously visited location using lidar data, especially when dealing with scans taken from both ground-based robots and aerial drones in challenging outdoor environments.

Not ideal if your primary need is 2D mapping, visual-only localization, or if you are working with environments that lack significant 3D structural information from lidar.

robot-localization autonomous-navigation 3d-mapping lidar-data-analysis geospatial-robotics
No Package No Dependents
Maintenance 10 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

26

Forks

3

Language

Python

License

Last pushed

Feb 05, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/csiro-robotics/HOTFormerLoc"

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