Shore Livecams: A Maritime Dataset for Deep Learning based Object Detection

Datasets of maritime objects are very important for training applications related to activity monitoring in locations around ports and shores. This paper introduces a maritime dataset for object detection, named as the shore livecam dataset. The dataset is a collection of high definition (HD), full high definition (FHD), and ultra high definition (UHD) images captured from live video feeds recorded across various port based areas of Germany. These images contain multiple instances of objects which are primarily classified into three different classes and annotated accordingly. The widely varying object sizes contribute to the uniqueness of this dataset whose content is thoroughly analysed and described. Finally, a selection of deep learning models is used on this dataset for evaluation. The dataset is available at https://github.com/HSU-ANT/shore_livecams/