liyih/HybridPoint

[ICME 2023, Oral] HybridPoint: Point cloud registration based on hybrid point sampling and matching

29
/ 100
Experimental

This project helps robotics engineers and 3D vision system developers precisely align multiple 3D scans or point clouds of an environment or object. It takes two partially overlapping point clouds as input and outputs a highly accurate transformation that maps one onto the other, ensuring consistent positioning. This is useful for tasks like 3D reconstruction, autonomous navigation, and object recognition.

No commits in the last 6 months.

Use this if you need to align 3D point cloud data with high accuracy and robustness, particularly in challenging real-world scenarios like those found in robotics or autonomous driving.

Not ideal if your primary goal is real-time processing with minimal computational overhead, as this method prioritizes accuracy over extreme speed.

3D-reconstruction robotics-navigation computer-vision LiDAR-processing autonomous-vehicles
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 6 / 25

How are scores calculated?

Stars

29

Forks

2

Language

Python

License

MIT

Last pushed

Mar 14, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/liyih/HybridPoint"

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