csiro-robotics/FactoFormer

[IEEE T-GRS 2024] The official repository for Journal Article “FactoFormer: Factorized Hyperspectral Transformers with Self-Supervised Pre-Training”, Accepted to IEEE Transactions on Geoscience and Remote Sensing, December 2023.

26
/ 100
Experimental

This project offers an advanced method for classifying land cover and objects in satellite imagery. It takes hyperspectral images—which capture light across many more spectral bands than a regular camera—and processes them to identify different materials or features on the Earth's surface. Geoscientists, environmental researchers, and remote sensing specialists will find this useful for precise mapping and analysis.

No commits in the last 6 months.

Use this if you need to accurately classify complex hyperspectral satellite images, especially when both the spectral 'fingerprint' and the spatial patterns of the land cover are important for differentiation.

Not ideal if you are working with standard RGB or multispectral images, or if your primary goal is not high-precision classification of intricate ground features.

remote-sensing hyperspectral-imaging land-cover-mapping geospatial-analysis earth-observation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 4 / 25

How are scores calculated?

Stars

22

Forks

1

Language

Python

License

Last pushed

Jun 22, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/csiro-robotics/FactoFormer"

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