2nd Workshop on Maritime Computer Vision (MaCVi)

Keynote speakers and topics

Kakani Katija

Unveiling FathomNet: Harnessing Machine Learning for Marine Biota Surveillance and Oceanic Stewardship

Kakani Katija, Principal Engineer at Monterey Bay Aquarium Research Institute

The ocean is experiencing unprecedented rapid change, and visually monitoring marine biota at the spatiotemporal scales needed for responsible stewardship is a formidable task. As baselines are sought by the research community, the volume and rate of this required data collection rapidly outpaces our abilities to process and analyze them. Recent advances in machine learning enables fast, sophisticated analysis of visual data, but have had limited success in the ocean due to lack of data standardization, insufficient formatting, and demand for large, labeled datasets. To address this need, we built FathomNet, an open-source image database that standardizes and aggregates expertly curated labeled data. FathomNet has been seeded with existing iconic and non-iconic imagery of marine animals, underwater equipment, debris, and other concepts, and allows for future contributions from distributed data sources. In addition to image data, FathomNet also aggregates machine learning models trained on this image resource and makes these models widely available through the FathomNet Model Zoo. As FathomNet continues to grow and incorporate more labeled data from the community, we can accelerate the processing of visual data to achieve a healthy and sustainable global ocean.


Innovation at Sea: Enhancing Unmanned Vehicles with Deep Learning Capabilities

Danilo Petrocelli, Senior Machine Learning Engineer, Maritime Robotics AS

The integration of Machine Learning / Deep Learning with autonomous maritime robotics is set to reshape the capabilities of Unmanned Aerial Vehicles (UAVs) and Unmanned Surface Vehicles (USVs) in maritime environments. Throughout our presentation, we will offer an overview of Maritime Robotics’ ongoing R&D projects and cutting-edge products, showcasing state-of-the-art Machine Learning and Deep Learning solutions. We will highlight our latest solution designed to enhance situational awareness for remote USV operators. Furthermore, we will delve into the challenges associated with deploying these state-of-the-art Deep Learning solutions in real-world applications and discuss effective mitigations.


MEMS-scanner-based laser projection system for maritime augmented reality and LIDAR camera for underwater scenarios

Shanshan Gu-Stoppel, Head of Group Optical Systems, Fraunhofer ISIT

Devices based on MEMS technology have great potential for maritime applications thanks to their small volume, low weight and low power consumption. In this presentation, two MEMS-based systems for maritime applications will be discussed. As part of a funded research project ("MEMS scanner-based laser projection system for maritime augmented reality"), a projection module was developed that combines the biaxial MEMS scanner, an RGB laser beam combiner and the electronics for readout and control. The aim was to develop a smart window in the form of a MEMS scanner-based laser projection system for maritime augmented reality, which offers the possibility of displaying safety-relevant information from navigation. A second MEMS system is the AI-supported LiDAR camera with potential to support underwater operations. The development of a LIDAR system combining MEMS deflection unit, detector, ultra-low divergence PCSEL laser and the AI approach is reported.