llmbev/talk2bev

Talk2BEV: Language-Enhanced Bird's Eye View Maps (ICRA'24)

38
/ 100
Emerging

This project helps autonomous vehicle engineers interpret complex driving scenes using natural language. It takes bird's-eye view (BEV) maps, which are common in self-driving systems, and allows you to ask questions about them in plain English. The output is answers that help understand traffic actor intentions and make decisions based on visual information. Autonomous vehicle developers and safety engineers would use this to enhance scene understanding.

119 stars. No commits in the last 6 months.

Use this if you need to perform advanced visual and spatial reasoning on autonomous driving BEV maps using natural language queries, without needing to train for specific object categories.

Not ideal if your primary need is for a system that relies solely on predefined, closed sets of object categories and driving scenarios without natural language interaction.

autonomous-driving vehicle-perception scene-understanding robotics traffic-analysis
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

119

Forks

11

Language

Python

License

BSD-3-Clause

Last pushed

Nov 04, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/llmbev/talk2bev"

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