wenyuqing/panacea

[CVPR2024] Official Repository of Paper "Panacea: Panoramic and Controllable Video Generation for Autonomous Driving"

37
/ 100
Emerging

This project helps automotive engineers and researchers create realistic, controllable driving videos from Bird's-Eye-View (BEV) layout sequences and other conditions. You input structured BEV data (like bounding boxes, object depths, road maps, and camera poses) along with optional conditions like weather or time of day. The output is a high-fidelity, multi-view video that simulates diverse driving scenarios, even rare or extreme ones. It's used by professionals developing and testing autonomous driving systems.

254 stars. No commits in the last 6 months.

Use this if you need to generate synthetic, controllable video datasets to train and evaluate autonomous driving perception models, especially for rare or challenging scenarios.

Not ideal if you are looking to process existing video footage or if you don't have access to BEV layout data for your desired scenarios.

autonomous-driving synthetic-data-generation sensor-simulation AI-model-training scenario-testing
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 11 / 25

How are scores calculated?

Stars

254

Forks

15

Language

Python

License

Apache-2.0

Last pushed

Aug 15, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/wenyuqing/panacea"

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