Marine Obstacle Detection Dataset and evaluation code (NEW!)
This is home page of MODD dataset. This dataset contains marine videos, captured by unmanned surface vehicle (USV). The challenge, posed by this dataset, is to segment each image into three natural regions: the sky, the shore and the sea, and furthermore, detect obstacles in the sea area.
We also provide four additional non-annotated videos. Download them via this link. They are provided solely for qualitative analysis.
If you use this dataset, please cite:
 M. Kristan, V. Sulic Kenk, S. Kovacic and J. Pers, "Fast Image-Based Obstacle
Detection From Unmanned Surface Vehicles," in IEEE Transactions on Cybernetics,
vol. 46, no. 3, pp. 641-654, March 2016. doi: 10.1109/TCYB.2015.2412251
 Matej Kristan, Janez Pers,
Vildana Sulic, Stanislav Kovacic. A Graphical Model for Rapid
Obstacle Image-map Estimation from Unmanned Surface Vehicles, ACCV
2014, Singapore, November 2014.
The papers and the corresponding code are available here.
We separately provide evaluation code, which can be used to reproduce our results, or evaluate your methods. The code is complete evaluation and processing (algorithm) package, which includes the algorithm described in the papers above. You can download it from this link. Before downloading, please
read the associated README file
The videos below were generated from images and ground truth annotations in the MODD dataset. They are provided as illustration what type of data is provided with MODD, and show four types
of annotations: large obstacles (straddling the edge of the sea), small obstacles (totally enclosed within the sea surface), edge of the sea (line/polygon), and glitter (only the last
two videos, which do not contain any obstacles).