YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
LSeg for Remote Sensing: Language-Driven Earth Observation Segmentation
This model adapts the LSeg (Language-Driven Semantic Segmentation) framework specifically for Remote Sensing (RS) and Satellite Imagery.
Remote sensing images introduce unique spatial complexities—such as scale variations, high object density, off-nadir angles, and multi-spectral distributions—making open-vocabulary segmentation particularly challenging. This project combines domain-specific visual backbones with open-vocabulary text embeddings to enable flexible, text-prompted segmentation on Earth observation data.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support