Voxel51/VisDrone2019-DET|无人机视觉数据集|目标检测数据集

---
annotations_creators: []
language: en
license: cc-by-sa-3.0
task_categories:
- object-detection
task_ids: []
pretty_name: VisDrone2019-DET
tags:
- fiftyone
- image
- object-detection
dataset_summary: >


This is a [FiftyOne](https://github.com/voxel51/fiftyone) dataset with 8629
samples.

## Installation

If you haven't already, install FiftyOne:

```bash

pip install -U fiftyone

```

## Usage

```python

import fiftyone as fo

import fiftyone.utils.huggingface as fouh

# Load the dataset
dataset = fouh.load_from_hub("Voxel51/VisDrone2019-DET")

## Or just load the first 1000 samples
## dataset = fouh.load_from_hub("Voxel51/VisDrone2019-DET", max_samples=1000)

# Launch the App

session = fo.launch_app(dataset)

```
size_categories:
- 1K<n<10K
---

# Dataset Card for VisDrone2019-DET


This is a [FiftyOne](https://github.com/voxel51/fiftyone) version of the VisDrone2019-DET dataset with 8629 samples.

## Installation

If you haven't already, install FiftyOne:

```bash
pip install -U fiftyone
```

## Usage

```python
import fiftyone as fo
import fiftyone.utils.huggingface as fouh

# Load the dataset
# Note: other available arguments include 'max_samples', 'persistent`, 'overwrite' etc
dataset = fouh.load_from_hub("Voxel51/VisDrone2019-DET")

# Launch the App
session = fo.launch_app(dataset)
```

## Dataset Details

### Dataset Description

<!-- Provide a longer summary of what this dataset is. -->

- **Curated by:** AISKYEYE team at the Lab of Machine Learning and Data Mining, Tianjin University, China
- **Language(s) (NLP):** en
- **License:** cc-by-sa-3.0

### Dataset Sources
<!-- Provide the basic links for the dataset. -->

- **Repository:** https://github.com/VisDrone/VisDrone-Dataset
- **Paper:** [Detection and Tracking Meet Drones Challenge](https://arxiv.org/abs/2001.06303)

## Dataset Structure

```plaintext
Name: VisDrone2019-DET
Media type: image
Num samples: 8629
Persistent: False
Tags: []
Sample fields:
id: fiftyone.core.fields.ObjectIdField
filepath: fiftyone.core.fields.StringField
tags: fiftyone.core.fields.ListField(fiftyone.core.fields.StringField)
metadata: fiftyone.core.fields.EmbeddedDocumentField(fiftyone.core.metadata.ImageMetadata)
ground_truth: fiftyone.core.fields.EmbeddedDocumentField(fiftyone.core.labels.Detections)
```

The dataset has 3 splits: "train", "val", and "test". Samples are tagged with their split.

## Dataset Creation

Created by the AISKYEYE team at the Lab of Machine Learning and Data Mining, Tianjin University, China.

### Source Data

#### Who are the source data producers?

The VisDrone Dataset is a large-scale benchmark created by the AISKYEYE team at the Lab of Machine Learning and Data Mining, Tianjin University, China. It contains carefully annotated ground truth data for various computer vision tasks related to drone-based image and video analysis.

#### Personal and Sensitive Information

The authors of the dataset have done their best to exclude identifiable information from the data to protect privacy. If you find your vehicle or personal information in this dataset, please [contact them]([e-mail]) and they will remove the corresponding information from their dataset. They are not responsible for any actual or potential harm as the result of using this dataset.

<!-- State whether the dataset contains data that might be considered personal, sensitive, or private (e.g., data that reveals addresses, uniquely identifiable names or aliases, racial or ethnic origins, sexual orientations, religious beliefs, political opinions, financial or health data, etc.). If efforts were made to anonymize the data, describe the anonymization process. -->

## Citation

**BibTeX:**

```bibtex
@ARTICLE{9573394,
author={Zhu, Pengfei and Wen, Longyin and Du, Dawei and Bian, Xiao and Fan, Heng and Hu, Qinghua and Ling, Haibin},
journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
title={Detection and Tracking Meet Drones Challenge},
year={2021},
volume={},
number={},
pages={1-1},
doi={10.1109/TPAMI.2021.3119563}}
```

## Copyright Information

The copyright of the [VisDrone dataset](https://github.com/VisDrone/VisDrone-Dataset) is reserved by the AISKYEYE team at Lab of Machine Learning and Data Mining, Tianjin University, China. The dataset described on this page is distributed under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License, which implies that you must:
(1) attribute the work as specified by the original authors;
(2) may not use this work for commercial purposes ;
(3) if you alter, transform, or build upon this work, you may distribute the resulting work only under the same license.
The dataset is provided “as it is” and we are not responsible for any subsequence from using this dataset.

原始信息汇总

数据集概述

名称: VisDrone2019-DET
样本数量: 8629
语言: 英语
许可证: cc-by-sa-3.0
任务类别: 目标检测
媒体类型: 图像
数据集创建者: AISKYEYE团队，天津大学机器学习和数据挖掘实验室
数据集来源:

• 仓库: https://github.com/VisDrone/VisDrone-Dataset
• 论文: Detection and Tracking Meet Drones Challenge

数据集结构

• 样本字段:

  • id: fiftyone.core.fields.ObjectIdField
  • filepath: fiftyone.core.fields.StringField
  • tags: fiftyone.core.fields.ListField(fiftyone.core.fields.StringField)
  • metadata: fiftyone.core.fields.EmbeddedDocumentField(fiftyone.core.metadata.ImageMetadata)
  • ground_truth: fiftyone.core.fields.EmbeddedDocumentField(fiftyone.core.labels.Detections)

• 数据集分割: 训练集、验证集、测试集

数据集创建

• 源数据生产者: AISKYEYE团队，天津大学机器学习和数据挖掘实验室
• 个人和敏感信息: 数据集作者已尽力排除可识别信息以保护隐私。如发现个人或车辆信息，请联系作者进行移除。

引用信息

bibtex @ARTICLE{9573394, author={Zhu, Pengfei and Wen, Longyin and Du, Dawei and Bian, Xiao and Fan, Heng and Hu, Qinghua and Ling, Haibin}, journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, title={Detection and Tracking Meet Drones Challenge}, year={2021}, volume={}, number={}, pages={1-1}, doi={10.1109/TPAMI.2021.3119563}}

版权信息

• 版权所有者: AISKYEYE团队，天津大学机器学习和数据挖掘实验室
• 许可证详情: 创意共享署名-非商业性使用-相同方式共享3.0许可证

以上信息基于提供的数据集详情页面README文件内容整理。

AI搜集汇总

数据集介绍

构建方式

VisDrone2019-DET数据集由AISKYEYE团队在天津大学机器学习与数据挖掘实验室精心构建，包含8629个样本。该数据集通过无人机拍摄的图像进行标注，涵盖了多种计算机视觉任务，如目标检测。数据集分为训练、验证和测试三个部分，每个样本均附有详细的标注信息，确保了数据的高质量和实用性。

特点

VisDrone2019-DET数据集的主要特点在于其大规模和高精度。数据集包含8629张图像，每张图像均经过精细标注，适用于目标检测任务。此外，数据集的多样性和复杂性使其成为评估和训练计算机视觉模型的理想选择。数据集还遵循CC BY-SA 3.0许可协议，确保了使用的灵活性和合法性。

使用方法

使用VisDrone2019-DET数据集，首先需要安装FiftyOne库，通过以下命令进行安装：`pip install -U fiftyone`。随后，可以使用Python代码加载数据集，例如：`dataset = fouh.load_from_hub("Voxel51/VisDrone2019-DET")`。加载后，可以通过FiftyOne的应用程序界面进行数据集的可视化和分析，进一步支持目标检测模型的训练和评估。

背景与挑战

背景概述

VisDrone2019-DET数据集由天津大学机器学习与数据挖掘实验室的AISKYEYE团队创建，旨在为无人机图像分析中的目标检测任务提供一个大规模的基准。该数据集包含8629个样本，涵盖了多种复杂的场景和目标类型，为计算机视觉领域的研究提供了丰富的资源。其核心研究问题是如何在无人机视角下准确地检测和识别目标，这对于无人机在实际应用中的导航、监控和搜索救援等任务至关重要。该数据集的发布不仅推动了目标检测技术的发展，也为相关领域的研究提供了宝贵的数据支持。

当前挑战

VisDrone2019-DET数据集在构建过程中面临了多重挑战。首先，无人机视角下的图像具有高度的动态性和复杂性，导致目标检测的难度增加。其次，数据集的标注工作需要高精度的标注，以确保训练模型的准确性。此外，数据集中可能包含敏感信息，如车辆和个人身份，需要进行严格的隐私保护措施。这些挑战不仅影响了数据集的质量，也对后续的研究和应用提出了更高的要求。

常用场景

经典使用场景

在计算机视觉领域，Voxel51/VisDrone2019-DET数据集以其丰富的无人机视角图像和精确的对象检测标注而著称。该数据集的经典使用场景主要集中在无人机图像中的目标检测任务，研究人员利用这些标注数据训练和评估各种深度学习模型，以提高无人机在复杂环境中的目标识别能力。

解决学术问题

Voxel51/VisDrone2019-DET数据集解决了无人机视角下目标检测的学术研究问题。通过提供高质量的标注数据，该数据集帮助研究人员克服了无人机图像中常见的遮挡、光照变化和视角多样性等挑战，推动了目标检测算法在实际应用中的性能提升。

衍生相关工作

基于Voxel51/VisDrone2019-DET数据集，许多相关工作得以展开，包括但不限于改进的目标检测算法、多目标跟踪技术以及无人机图像的实时处理系统。这些研究不仅提升了无人机在复杂环境中的感知能力，还为相关领域的技术进步提供了坚实的基础。