DDOS: The Drone Depth and Obstacle Segmentation Dataset

Overview

The Drone Depth and Obstacle Segmentation (DDOS) dataset is a collection of aerial images captured by drones, with corresponding depth maps, and annotated with pixel-wise segmentation masks. DDOS is designed to facilitate research and development in the field of computer vision, specifically addressing tasks related to depth estimation and obstacle segmentation from aerial imagery.

For a detailed understanding of DDOS, please refer to the associated research paper: DDOS: The Drone Depth and Obstacle Segmentation Dataset.

Please Note: The final documentation for this dataset has not been finalized yet. Therefore, the information provided here is subject to change. This README serves as a preliminary guide to give you an understanding of the dataset's structure and contents.

Data Structure

DDOS is organised as follows:

Data Splits:
- Train: Contains training data for model training.
- Validation: Contains data for validation during model development.
- Test: Contains data for the final evaluation of the trained model.
Environments:
- Neighbourhood: Contains data captured in urban and residential environments.
- Park: Contains data captured in park and natural environments.
Flights:
- Each flight is represented by a unique flight ID and is contained within the corresponding environment directory.
Data for Each Flight:
- Depth: Contains depth maps representing the distance of objects from the camera.
- Scene Flow: Contains scene flow data representing the 3D motion of objects.
- Image: Contains RGB images captured by the drone camera.
- Segmentation: Contains pixel-wise segmentation masks for obstacle segmentation.
- Surface Normal: Contains surface normal maps representing the orientation of object surfaces.

Additional Information

Class Mapping: The segmentation masks use the following class labels for obstacle segmentation:

CLASS_MAPPING = {
    'ultra_thin': 255,
    'thin_structures': 240,
    'small_mesh': 220,
    'large_mesh': 200,
    'trees': 180,
    'buildings': 160,
    'vehicles': 140,
    'people': 120,
    'animals': 100,
    'other': 80
}

Metadata: The dataset contains metadata, such as GPS coordinates, altitude, and camera parameters, which provide valuable contextual information about each flight.

Dataset Usage

Data Loading To load and utilize the DDOS dataset for your machine learning projects, you can refer to the official PyTorch data loading tutorial for beginners: PyTorch Data Loading Tutorial This tutorial will guide you through the process of loading data, creating data loaders, and preparing the dataset for training or evaluation using PyTorch.
Disclaimer: While every effort has been made to ensure the accuracy and quality of the DDOS dataset, the dataset owner and contributors cannot be held responsible for any inaccuracies or errors in the data. Users are advised to exercise caution and due diligence when using the dataset for their research or applications.

License

DDOS is openly licensed under CC BY-NC 4.0.

Citation

If you use DDOS in your research or projects, please cite our paper:

@article{kolbeinsson2023ddos,
  title={{DDOS}: The Drone Depth and Obstacle Segmentation Dataset},
  author={Benedikt Kolbeinsson and Krystian Mikolajczyk},
  journal={arXiv preprint arXiv:2312.12494},
  year={2023}
}