BallNet: A UAV Collision Avoidance Dataset

BallNet

Dataset for drones

BallNet, a dataset of collisions that aims to provide
a base for training new Machine Learning algorithms
that handle the problem of collisions with high efficiency
and robustness. We show that using this dataset
is easy to construct new neural network models and test them.

If you use this dataset or scripts to handle your video dataset, please cite:

@InProceedings{colanet2020,author="Dário Pedro, João Carvalho, José Fonseca and Andrá Mora",
title="Collision Avoidance on Unmanned Aerial Vehicles Using Neural Network Pipelines and Flow Clustering Techniques",
URL="https://www.mdpi.com/2072-4292/13/13/2643", DOI="10.3390/rs13132643",
journal="Remote Sensing", year="2021", number="13",
isbn="2072-4292}
"
More information in github!