Real-time embedded reconstruction of dynamic objects for a 3D maritime situational awareness picture

Abstract

Monitoring maritime infrastructures is an important part of maritime safety and security. To best assess the security status of these facilities, detailed information should be made available to stakeholders, such as port authorities, law enforcement agencies and emergency services in a concise and easily understandable format. In this work, we propose a novel real-time 3D reconstruction framework for enhancing maritime situational awareness. We introduce and verify a pipeline prototype for dynamic 3D reconstruction of maritime objects using a static observer and stereoscopic cameras on an GPU-accelerated embedded device. Our pipeline runs with approx. 6Hz on a Nvidia Jetson Xavier AGX embedded system and is verified using a simulated dataset of a harbor basin.