humansensinglab/AGenDA

[ICCV 2025] Adapting Vehicle Detectors for Aerial Imagery to Unseen Domains with Weak Supervision

25
/ 100
Experimental

This project helps refine vehicle detection models used with drone or satellite imagery, particularly when the conditions or locations in new images are different from those the model was originally trained on. It takes existing aerial images (some with weak labels) and an initial vehicle detection model, then outputs an improved model better suited for diverse, new environments. Analysts in aerial surveillance or urban planning who rely on automated vehicle counts will find this useful.

Use this if you need to reliably count or detect vehicles in aerial imagery from a wide range of geographical locations or weather conditions, even if your existing models struggle with these new scenarios.

Not ideal if you are looking for a pre-trained, ready-to-use vehicle detection model without needing to adapt it to new or challenging aerial imaging environments.

aerial-surveillance urban-planning traffic-monitoring drone-imagery-analysis remote-sensing
No License No Package No Dependents
Maintenance 6 / 25
Adoption 5 / 25
Maturity 7 / 25
Community 7 / 25

How are scores calculated?

Stars

11

Forks

1

Language

Python

License

Last pushed

Dec 01, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/humansensinglab/AGenDA"

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