EPFL-ENAC/TOPO-DataGen

[CVPR'22] TOPO-DataGen: an open and scalable aerial synthetic data generation workflow

26
/ 100
Experimental

This workflow helps urban planners, environmental scientists, or drone operators create diverse synthetic visual data from existing geospatial information. You input common geo-data like orthophotos, digital terrain models, or classified point clouds, and it generates synthetic RGB images, 3D scene coordinates, and semantic labels. This is ideal for anyone needing realistic aerial imagery and associated ground truth for tasks like training AI models.

No commits in the last 6 months.

Use this if you need to generate large, diverse datasets of synthetic aerial imagery with precise 3D geometry, semantic labels, and camera poses for a specific geographic area.

Not ideal if you don't have existing geo-data (like orthophotos or DTMs) for your target area, or if you only need real-world images without synthetic augmentation.

aerial-mapping geospatial-analysis remote-sensing urban-planning drone-operations
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 3 / 25

How are scores calculated?

Stars

35

Forks

1

Language

Jupyter Notebook

License

MIT

Last pushed

Dec 02, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/EPFL-ENAC/TOPO-DataGen"

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