The German Outdoor and Offroad Dataset (GOOSE) is a modern dataset specification and accompanying off-road datasets. The focus is on unstructured off-road environments as well as on a broad support for different platforms and applications in the fields of mobile robotics and deep learning.
This repository contains code to process and visualize data and to run benchmarks on different baseline methods. It is also used to track issues of the GOOSE and GOOSE-Ex datasets, the database, website, etc, so feel free to open an issue if anything is not working as expected.
The data structure and more in-depth information about the format can be found int the documentation. The data is divided into 3 splits: train, test and validation. Labeled data is available for train and validation splits.
It can be downloaded from our webpage.
In scripts you can find some sample scripts to directly download and unpack the 2D data.
Under the folder common some general configuration files and utils such as color maps can be found.
For more specific tools regarding training and data handling, have a look at the image_processing and pointcloud_processing subfolders.
Please cite us if this data is useful for you work:
@article{goose-dataset,
author = {Peter Mortimer and Raphael Hagmanns and Miguel Granero
and Thorsten Luettel and Janko Petereit and Hans-Joachim Wuensche},
title = {The GOOSE Dataset for Perception in Unstructured Environments},
url={https://arxiv.org/abs/2310.16788},
conference={2024 IEEE International Conference on Robotics and Automation (ICRA)}
year = 2024
}
@article{goose-ex-dataset,
author = {Raphael Hagmanns and Peter Mortimer and Miguel Granero
and Thorsten Luettel and Janko Petereit},
title = {Excavating in the Wild: The GOOSE-Ex Dataset for Semantic Segmentation},
url={},
conference={TBA}
year = 2024
}- This repository is licensed under the MIT License.
- The data is published under the CC BY-SA 4.0 License.
GOOSE is a project of Fraunhofer IOSB, UniBW Munich and University of Koblenz.