| task_categories: | |
| - image-segmentation | |
| This repository contains the dataset used in the paper [Bridging the Gap between Labeled and Unlabeled Data via Unified Flow with Feature Memory Bank](https://huggingface.co/papers/2608.16681). The dataset consists of remote sensing images and their segmentation masks for semi-supervised semantic segmentation. Code: [https://github.com/wangshanwen001/RS-UFFM](https://github.com/wangshanwen001/RS-UFFM). | |
| # Acknowledgments | |
| We sincerely thank the authors of [SAM3](https://github.com/facebookresearch/sam3) and [UniMatch](https://github.com/LiheYoung/UniMatch) for their excellent open‑source work, and we also thank the contributors of the publicly available [DeepGlobe](https://www.kaggle.com/datasets/balraj98/deepglobe-land-cover-classification-dataset?spm=5176.28103460.0.0.96a02988pIA0pI) and [Potsdam](https://www.isprs.org/resources/datasets/benchmarks/UrbanSemLab/2d-sem-label-potsdam.aspx?utm_source=chatgpt.com) datasets. Please follow the licenses and terms of the original models and datasets. |