Add dataset card with metadata, paper, and GitHub links

README.md

@@ -1,3 +1,66 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+task_categories:
+- image-segmentation
+tags:
+- remote-sensing
+- open-vocabulary
+---
+
+# OVRSIS95K
+
+This repository contains the **OVRSIS95K** dataset, which is the foundation training dataset for **OVRSISBenchV2** presented in the paper [Towards Realistic Open-Vocabulary Remote Sensing Segmentation: Benchmark and Baseline](https://huggingface.co/papers/2604.15652).
+
+Official GitHub Repository: [LiBingyu01/Pi-Seg](https://github.com/LiBingyu01/Pi-Seg)
+
+## Dataset Highlights
+
+- **~95K image-mask pairs** with high-quality annotations.
+- **35 common semantic categories** with balanced class distribution.
+- **5 scene domains**: town, industrial, forest, waterfront, and wasteland.
+- **Semi-automated annotation pipeline** to ensure both scalability and quality.
+
+## Dataset Structure
+
+The dataset contains remote sensing images and corresponding segmentation annotations. When downloading and preparing the dataset, organize the files as follows:
+
+```
+datasets/
+└── OVRSIS95K/
+    ├── train/
+    │   ├── images/
+    │   └── annotations/
+    └── val/
+        ├── images/
+        └── annotations/
+```
+
+## Related Datasets
+
+The OVRSISBenchV2 benchmark evaluates open-world generalization across multiple sets:
+* **OVRSIS95K (Training)**: [Hugging Face](https://huggingface.co/datasets/kkk2026/OVRSIS95K)
+* **OVRSISBenchV2 (OVRSIS test)**: [Hugging Face](https://huggingface.co/datasets/kkk2026/OVRSISBenchtest)
+* **OVRSISBenchV2 (3 downstream tasks)**: [Hugging Face](https://huggingface.co/datasets/kkk2026/OVRSISBenchV2o3)
+
+## Citation
+
+If you find this dataset or the associated benchmark/baseline useful, please cite:
+
+```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}
+}
+```