bhavyabafna/UEMM-Air

---
license: apache-2.0
pretty_name: UEMM-Air
size_categories:
- 100B<n<1T
---
# UEMM-Air: Enable UAVs to Undertake More Multi-modal Tasks

## 📋 Table of Contents

- [News](#news)
- [Introduction](#introduction)
- [Download the UEMM-Air 📂](#download-the-uemm-air-)
- [Multi-modality Images](#multi-modality-images)
- [Object Detection](#object-detection)
- [Instance Segmentation](#instance-segmentation)
- [Referring Expression Segmentation](#referring-expression-segmentation)
- [Image-Text Retrieval](#image-text-retrieval)
- [License 🚨](#license-)
- [Citation 🎈](#citation)
- [Contact ✉](#contact-)

## News
- **2025/1/20**: We have open-sourced the dataset generation system, which can be found in the [AirNavigation](https://github.com/1e12Leon/AirNavigation).
- **2024/12/11**: Welcome to UEMM-Air! Dataset is open-sourced at this repository.

## Introduction

We present a large-scale synthetic drone vision dataset with 6 paired multimodal streams (120k+ sequences) and 4D task versatility ,
enabling comprehensive research in perception, navigation, and autonomy.
Built on Unreal Engine, it offers photorealistic aerial scenarios with precise physics, diverse environmental variations, and pixel-perfect annotations.
The paired modalities facilitate cross-modal learning and domain adaptation studies, while the multi-task support (detection, segmentation, retrieval, cross-modality understanding)
encourages holistic perception modeling. Its synthetic nature ensures scalability, reproducibility, and rare-event coverage, addressing critical gaps in real-world drone datasets. This work establishes a new benchmark for robust, generalizable vision systems in complex aerial environments.

## Download the UEMM-Air 📂

* 🤗[Hugging Face](https://huggingface.co/datasets/1e12Leon/UEMM-Air)

### Multi-modality Images
* [BaiduYun](https://pan.baidu.com/s/1AgrehM3Bs-aiVLVrdswWeQ?pwd=xcpe)

### Object Detection
* [BaiduYun](https://pan.baidu.com/s/1bkG3G3nUre65yk0XjeaQ5w?pwd=a3qt)

### Instance Segmentation
* [BaiduYun](https://pan.baidu.com/s/1TEwa8NrmbDK_Vd_zpysHug?pwd=y1f4)

### Referring Expression Segmentation
* [BaiduYun](https://pan.baidu.com/s/1hO5h2UdYwxJrLmk4oStupg?pwd=wqxi)

### Image-Text Retrieval
* [BaiduYun](https://pan.baidu.com/s/1O-U84fhqsJruyEV-UDKx8w?pwd=jppd)

## License 🚨
This dataset is licensed under the [Creative Commons Attribution-NonCommercial 4.0 International License (CC-BY-NC 4.0)](https://creativecommons.org/licenses/by-nc/4.0/).

By downloading or using the Dataset, as a Licensee I/we understand, acknowledge, and hereby agree to all the terms of use. This dataset is provided "as is" and without any warranty of any kind, express or implied. The authors and their affiliated institutions are not responsible for any errors or omissions in the dataset, or for the results obtained from the use of the dataset. **The dataset is intended for academic research purposes only, and not for any commercial or other purposes.** The users of the dataset agree to acknowledge the source of the dataset and cite the relevant papers in any publications or presentations that use the dataset. The users of the dataset also agree to respect the intellectual property rights of the original data owners.

## Citation🎈

```bibtex
@misc{yao2025uemmair,
title={UEMM-Air: Make Unmanned Aerial Vehicles Perform More Multi-modal Tasks},
author={Liang Yao and Fan Liu and Shengxiang Xu and Chuanyi Zhang and Xing Ma and Jianyu Jiang and Zequan Wang and Shimin Di and Jun Zhou},
year={2025},
eprint={2406.06230},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2406.06230},
}
```

## Contact ✉
Please Contact [e-mail].

---
许可证：Apache 2.0
数据集展示名称：UEMM-Air
数据集规模区间：1000亿 < 样本量 < 1万亿
---
# UEMM-Air：赋能无人机（Unmanned Aerial Vehicle, UAV）开展更多多模态任务

## 📋 目录
- [最新动态](#news)
- [数据集简介](#introduction)
- [下载UEMM-Air 📂](#download-the-uemm-air-)
- [多模态图像](#multi-modality-images)
- [目标检测](#object-detection)
- [实例分割](#instance-segmentation)
- [指代表达式分割（Referring Expression Segmentation）](#referring-expression-segmentation)
- [图像-文本检索](#image-text-retrieval)
- [许可证 🚨](#license-)
- [引用 🎈](#citation)
- [联系方式 ✉](#contact-)

## 最新动态
- **2025/1/20**：我们已开源数据集生成系统，可在[AirNavigation](https://github.com/1e12Leon/AirNavigation)仓库获取。
- **2024/12/11**：欢迎使用UEMM-Air！本数据集已在此仓库开源。


## 数据集简介
我们发布了一款大规模合成无人机视觉数据集，包含6组配对多模态流（12万+序列）与4类任务泛用性，可支撑感知、导航与自主决策领域的全面研究。该数据集基于虚幻引擎（Unreal Engine）构建，提供具备精确物理模拟、多样化环境变化与像素级精准标注的逼真航拍场景。配对模态数据可支持跨模态学习与域自适应研究，而多任务支持（目标检测、实例分割、指代表达式分割、图像-文本检索）则可推动全景感知建模的研究。其合成属性确保了数据集的可扩展性、可复现性与罕见场景覆盖能力，填补了真实世界无人机数据集的关键空白。本工作为复杂航拍环境下的鲁棒、通用视觉系统建立了全新基准。

## 下载UEMM-Air 📂

* 🤗[Hugging Face](https://huggingface.co/datasets/1e12Leon/UEMM-Air)

### 多模态图像
* [百度网盘](https://pan.baidu.com/s/1AgrehM3Bs-aiVLVrdswWeQ?pwd=xcpe)

### 目标检测
* [百度网盘](https://pan.baidu.com/s/1bkG3G3nUre65yk0XjeaQ5w?pwd=a3qt)

### 实例分割
* [百度网盘](https://pan.baidu.com/s/1TEwa8NrmbDK_Vd_zpysHug?pwd=y1f4)

### 指代表达式分割（Referring Expression Segmentation）
* [百度网盘](https://pan.baidu.com/s/1hO5h2UdYwxJrLmk4oStupg?pwd=wqxi)

### 图像-文本检索
* [百度网盘](https://pan.baidu.com/s/1O-U84fhqsJruyEV-UDKx8w?pwd=jppd)

## 许可证 🚨
本数据集采用[知识共享署名-非商业性使用4.0国际许可协议（Creative Commons Attribution-NonCommercial 4.0 International License, CC-BY-NC 4.0）](https://creativecommons.org/licenses/by-nc/4.0/)授权。

下载或使用本数据集即视为被许可方理解、确认并同意遵守全部使用条款。本数据集按“现状”提供，不附带任何明示或暗示的担保。作者及其所属机构不对数据集中的任何错误或疏漏，以及因使用本数据集所获得的结果承担责任。**本数据集仅用于学术研究目的，不得用于任何商业或其他用途。**数据集使用者应在使用本数据集的任何出版物或演示中注明数据集来源并引用相关论文。数据集使用者还应尊重原始数据所有者的知识产权。

## 引用🎈

bibtex
@misc{yao2025uemmair,
title={UEMM-Air: Make Unmanned Aerial Vehicles Perform More Multi-modal Tasks},
author={Liang Yao and Fan Liu and Shengxiang Xu and Chuanyi Zhang and Xing Ma and Jianyu Jiang and Zequan Wang and Shimin Di and Jun Zhou},
year={2025},
eprint={2406.06230},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2406.06230},
}

## 联系方式 ✉
请联系邮箱 [e-mail]。