Add dataset card, task categories and links to paper and code

README.md

@@ -1,3 +1,66 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+task_categories:
+- image-segmentation
+tags:
+- remote-sensing
+- open-vocabulary
+- earth-observation
+---
+
+# OVRSIS95K
+
+This repository contains the **OVRSIS95K** dataset, which serves as the foundational training dataset for the **OVRSISBenchV2** benchmark, introduced in the paper [Towards Realistic Open-Vocabulary Remote Sensing Segmentation: Benchmark and Baseline](https://huggingface.co/papers/2604.15652).
+
+- **GitHub Repository**: [LiBingyu01/Pi-Seg](https://github.com/LiBingyu01/Pi-Seg)
+- **Paper**: [arXiv:2604.15652](https://huggingface.co/papers/2604.15652)
+
+---
+
+## Dataset Description
+
+**OVRSIS95K** is a large-scale, balanced dataset of approximately 95,000 image-mask pairs covering 35 common semantic categories across diverse remote sensing scenes. It is designed to act as the core training foundation for evaluating open-world generalization in Remote Sensing Image Segmentation (OVRSIS).
+
+### Key Features
+* **Size**: ~95,000 high-quality annotated image-mask pairs.
+* **Semantic Categories**: 35 balanced categories.
+* **Domains**: 5 diverse scene domains including *town*, *industrial*, *forest*, *waterfront*, and *wasteland*.
+
+### Dataset Structure
+To use this dataset with the official implementation, organize the directories as follows:
+
+```text
+datasets/
+└── OVRSIS95K/
+    ├── train/
+    │   ├── images/
+    │   └── annotations/
+    └── val/
+        ├── images/
+        └── annotations/
+```
+
+---
+
+## Citation
+
+If you find this dataset or the associated baseline code useful, please cite the following papers:
+
+```bibtex
+@article{li2026towards,
+  title={Towards Realistic Open-Vocabulary Remote Sensing Segmentation: Benchmark and Baseline},
+  author={Li, Bingyu and Huo, Tao and Dong, Haocheng and Zhang, Da and Zhao, Zhiyuan and Gao, Junyu and Li, Xuelong},
+  journal={arXiv preprint arXiv:2604.15652},
+  year={2026}
+}
+
+@inproceedings{li2026exploring,
+  title={Exploring efficient open-vocabulary segmentation in the remote sensing},
+  author={Li, Bingyu and Dong, Haocheng and Zhang, Da and Zhao, Zhiyuan and Sun, Hao and Gao, Junyu},
+  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
+  volume={40},
+  number={8},
+  pages={5982--5991},
+  year={2026}
+}
+```