banu4prasad/VisDrone-Dataset

VisDrone Dataset (YOLO Format)

Overview

The VisDrone dataset is a large-scale visual object detection and tracking benchmark captured by drones. Developed by the AISKYEYE team at Tianjin University, it aims to facilitate research in computer vision tasks such as object detection, object tracking, and crowd analysis in aerial imagery. The dataset consists of high-resolution images and videos collected using drones flying over urban and suburban environments across various cities in China. These scenes include pedestrians, vehicles, bicycles, and other common objects, captured under different lighting conditions, angles, and motion patterns.

The dataset has been modified to include only the image data and labels in YOLO format. The original annotation files have been removed, and object labels were converted using provided scripts(from Ultralytics) to be compatible with YOLO-based object detection models.

Dataset Details

• Classes:
  • 0: pedestrian
  • 1: people
  • 2: bicycle
  • 3: car
  • 4: van
  • 5: truck
  • 6: tricycle
  • 7: awning-tricycle
  • 8: bus
  • 9: motor
  • 10: others

Dataset Structure

The repository follows the standard Ultralytics YOLO structural requirements for object detection, where each split contains paired images and labels directories:

VisDrone-Dataset/
├── VisDrone2019-DET-train/
│   ├── images/  # .jpg drone frames
│   └── labels/  # .txt YOLO bounding box coordinates
├── VisDrone2019-DET-val/
│   ├── images/
│   └── labels/
├── VisDrone2019-DET-test-dev/
│   ├── images/
│   └── labels/
└── VisDrone2019-DET-test-challenge/
    └── images/

Citation

If you use this dataset in your research, please cite the original benchmark paper:

@inproceedings{wen2021detection,
  title={Detection, tracking, and counting meets drones in crowds: A benchmark},
  author={Wen, Longyin and Du, Dawei and Zhu, Pengfei and Hu, Qinghua and Wang, Qilong and Bo, Liefeng and Lyu, Siwei},
  booktitle={Proceedings of the IEEE/CVF conference on computer vision and pattern recognition},
  pages={7812--7821},
  year={2021}
}

@inproceedings{du2019visdrone,
  title={VisDrone-DET2019: The vision meets drone object detection in image challenge results},
  author={Du, Dawei and Zhu, Pengfei and Wen, Longyin and Bian, Xiao and Lin, Haibin and Hu, Qinghua and Peng, Tao and Zheng, Jiayu and Wang, Xinyao and Zhang, Yue and others},
  booktitle={Proceedings of the IEEE/CVF international conference on computer vision workshops},
  pages={0--0},
  year={2019}
}