Boats On Norderelbe at Kehrwieder
A 6D Pose Estimation Dataset for Marine Vessels
The Bonk-pose dataset is a novel, publicly available 6D pose estimation dataset for marine vessels, created by fusing monocular RGB images with Automatic Identification System (AIS) data. It addresses the gap in maritime datasets by providing fully annotated 3D bounding boxes for vessel pose estimation.
245 images shared between the 6D and 2D datasets.
Annotations are created by an automated data fusion approach,
making time-consuming human pose annotations unnecessary.
Annotations include vessel
centroid location in the real world, vessel dimensions and the directions of the vessel
coordinite system axes in the real world.
Intrinsic camera matrix is provided to project
this data into the image.
86.4% of vessels labeled on par with human annotations.
94.5%
of annotations are deemed of acceptable quality.
Object detection bounding boxes were created by humans. Vessels are annotated in five classes. The classes allow differentiation between vessels in the foreground, moving vessels and vessels part of the background.
The dataset will be made available at fabianholst.github.io/BONK-pose.
@dataset{bonk-pose, title = {Fusing Monocular RGB Images with AIS Data to Create a 6D Pose Estimation Dataset for Marine Vessels}, author = {Fabian Holst, Emre Gülsoylu, Simone Frintrop}, year = {2025}, url = {https://fabianholst.github.io/BONK-pose/} }