yjwong1999/RSGuidedDiffusion

[IEEE BigData 2024] Cross-City Building Instance Segmentation: From More Data to Diffusion-Augmentation

22
/ 100
Experimental

This project helps remote sensing analysts or urban planners generate realistic satellite-like images from basic land cover segmentation masks. You input existing segmentation maps (like outlines of buildings or land types), and it outputs new, synthetic satellite images that look real. This allows you to create more training data to improve the accuracy of models that automatically identify and segment buildings from satellite imagery, especially when real image data is scarce.

No commits in the last 6 months.

Use this if you need to create more diverse training data for remote sensing building segmentation tasks, especially when limited by the availability of actual satellite imagery, but you have segmentation masks.

Not ideal if you're looking to directly segment buildings from satellite images without using synthetic data, or if you don't have existing segmentation masks to guide the image generation process.

Remote Sensing Urban Planning Geographic Information Systems (GIS) Image Augmentation Building Extraction
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 4 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

8

Forks

Language

Python

License

MIT

Last pushed

Jun 12, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/yjwong1999/RSGuidedDiffusion"

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