jonathsch/lidar-synthesis

Official implementation of the ITSC 2023 paper "LiDAR View Synthesis for Robust Vehicle Navigation Without Expert Labels"

27
/ 100
Experimental

This project helps self-driving car developers and researchers create diverse training data for their navigation models. It takes existing LiDAR scans from safe driving scenarios and generates new, synthetic LiDAR point clouds from dangerous or hard-to-collect viewpoints. The output is augmented LiDAR data that improves the robustness of autonomous vehicle navigation systems.

No commits in the last 6 months.

Use this if you need to enhance the training datasets for self-driving car navigation models with synthetic LiDAR data, especially for hazardous driving scenarios.

Not ideal if you are looking for solutions that use only camera-based vision or require extensive expert human annotations for training data.

autonomous-driving LiDAR-simulation sensor-fusion data-augmentation vehicle-navigation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 5 / 25

How are scores calculated?

Stars

18

Forks

1

Language

Python

License

MIT

Last pushed

May 16, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/jonathsch/lidar-synthesis"

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