anxiangsir/urban_seg
Few-shot semantic segmentation framework for satellite and remote sensing imagery using a pre-trained vision foundation model.
Not currently ranked — collecting fresh signals.
star history
Urban Segmentation is a semantic segmentation framework for remote sensing imagery that leverages UNICOM, a vision foundation model pre-trained on 400 million images. It achieves competitive results with extreme data efficiency, requiring as few as 4 labeled satellite images for training. The project implements a streamlined pipeline for fine-tuning foundation models on overhead imagery segmentation tasks.
Frequently asked
- What is anxiangsir/urban_seg?
- Few-shot semantic segmentation framework for satellite and remote sensing imagery using a pre-trained vision foundation model.
- Is urban_seg open source?
- Yes — anxiangsir/urban_seg is open source, released under the Apache-2.0 license.
- What language is urban_seg written in?
- anxiangsir/urban_seg is primarily written in Python.
- How popular is urban_seg?
- anxiangsir/urban_seg has 473 stars on GitHub.
- Where can I find urban_seg?
- anxiangsir/urban_seg is on GitHub at https://github.com/anxiangsir/urban_seg.