BeamNG/impactgen
Python script and Lua extension using BeamNG.tech to generate low impact crash scenarios and ground truth data for imitation learning.
This tool helps researchers and engineers create a diverse dataset of vehicle crash scenarios for training AI models. It takes in vehicle models and crash parameters like impact speed and angle, then generates detailed image sequences of the crashes (both regular and semantically annotated) along with vehicle damage data. This is ideal for those developing computer vision systems for autonomous vehicles, insurance claim assessment, or accident reconstruction.
No commits in the last 6 months.
Use this if you need large-scale, high-quality synthetic data for training machine learning models that analyze vehicle impacts and damage.
Not ideal if you need to simulate complex, multi-vehicle pile-ups or highly customized crash environments not covered by the predefined scenarios.
Stars
19
Forks
5
Language
Python
License
MIT
Category
Last pushed
Apr 10, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/generative-ai/BeamNG/impactgen"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.