README.md

---
license: mit
datasets:
- JTRNEO/SynRS3D
language:
- en
base_model:
- facebook/dinov2-large
---

# SynRS3D: A Synthetic Dataset for Global 3D Semantic Understanding from Monocular Remote Sensing Imagery

**Authors:**  
[Jian Song](https://scholar.google.ch/citations?user=CgcMFJsAAAAJ&hl=zh-CN)<sup>1,2</sup>, [Hongruixuan Chen](https://scholar.google.ch/citations?user=XOk4Cf0AAAAJ&hl=zh-CN&oi=ao)<sup>1</sup>, [Weihao Xuan](https://weihaoxuan.com/)<sup>1,2</sup>, [Junshi Xia](https://scholar.google.com/citations?user=n1aKdTkAAAAJ&hl=en)<sup>2</sup>, [Naoto Yokoya](https://scholar.google.co.jp/citations?user=DJ2KOn8AAAAJ&hl=en)<sup>1,2</sup>

<sup>1</sup> The University of Tokyo  
<sup>2</sup> RIKEN AIP

**Conference:** Neural Information Processing Systems (Spotlight), 2024

For more details, please refer to our [paper](https://arxiv.org/pdf/2406.18151) and visit our GitHub [repository](https://github.com/JTRNEO/SynRS3D).

---

### Overview

**TL;DR:**  
We are excited to release two high-performing models for **height estimation** and **land cover mapping**. These models were trained on the SynRS3D dataset using our novel domain adaptation method, **RS3DAda**. 

- **Encoder:** Vision Transformer (ViT-L), pretrained with **DINOv2**  
- **Decoder:** [DPT](https://arxiv.org/abs/2103.13413), trained from scratch

These models excel in tasks involving large-scale global 3D semantic understanding from high-resolution remote sensing imagery. Feel free to integrate them into your projects for enhanced performance in related applications.

---

### How to Cite

If you find the RS3DAda model useful in your research, please consider citing:

```
@article{song2024synrs3d,
title={SynRS3D: A Synthetic Dataset for Global 3D Semantic Understanding from Monocular Remote Sensing Imagery},
author={Song, Jian and Chen, Hongruixuan and Xuan, Weihao and Xia, Junshi and Yokoya, Naoto},
journal={arXiv preprint arXiv:2406.18151},
year={2024}
}
```

---

### Contact

For any questions or feedback, please reach out via email at **song@ms.k.u-tokyo.ac.jp**.

We hope you enjoy using the pretrained RS3DAda models!