enginBozkurt/carla-training-data

Generating training data from the Carla driving simulator in the KITTI dataset format

45
/ 100
Emerging

This tool helps autonomous vehicle researchers and developers create synthetic training data for self-driving models. It takes simulated driving scenarios from the CARLA simulator and converts them into the standardized KITTI dataset format, which includes synchronized grayscale and color stereo images, Velodyne point clouds, GPS/IMU data, calibration information, and 3D object labels. This allows teams to expand their datasets without extensive real-world data collection.

109 stars. No commits in the last 6 months.

Use this if you need to generate diverse and labeled training data for autonomous driving perception tasks using a simulator and require the output in the KITTI benchmark format.

Not ideal if you are looking for a tool to collect or process real-world driving data, or if you don't use the CARLA simulator or the KITTI dataset format.

autonomous-driving computer-vision data-generation robotics vehicle-simulation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 20 / 25

How are scores calculated?

Stars

109

Forks

27

Language

Python

License

MIT

Last pushed

May 21, 2019

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/enginBozkurt/carla-training-data"

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