Doodleverse/segmentation_gym

A neural gym for training deep learning models to carry out geoscientific image segmentation. Works best with labels generated using https://github.com/Doodleverse/dash_doodler

44
/ 100
Emerging

This project helps geoscientists, environmental researchers, and remote sensing specialists analyze satellite, aerial, or other geoscientific images by automatically identifying and outlining specific features like coastlines, rivers, or land cover. You feed it your geoscientific images (like JPGs or PNGs) along with corresponding human-labeled examples, and it outputs a trained model capable of segmenting new, unlabeled images. This is for practitioners who need to efficiently process large volumes of imagery for environmental monitoring, mapping, or other geographical analyses.

No commits in the last 6 months.

Use this if you need to train robust deep learning models to automatically delineate features within geoscientific imagery, especially when working with multispectral or Earth Observation data.

Not ideal if your primary need is general image classification or object detection rather than precise pixel-level segmentation within geographical contexts.

geoscience remote-sensing environmental-monitoring image-analysis geographical-mapping
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 18 / 25

How are scores calculated?

Stars

53

Forks

14

Language

Python

License

MIT

Last pushed

Jul 22, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Doodleverse/segmentation_gym"

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