Each year maritime accidents occur at sea causing human casualties. Training facilities serve to reduce the risk of human error by allowing maritime teams to train safety procedures in cooperative real-size immersive simulators. However, they are expensive and only few maritime professionals have access to such simulators. Virtual Reality (VR) can provide a digital all-immersive learning environment at a reduced cost allowing for increased access. However, a key ingredient of what makes all-immersive physical simulators effective is that they allow for multiple participants to engage in cooperative social interaction. Social interaction which allows trainees to develop skills and competencies in navigating situational awareness essential for safety training. Social interaction requires social fidelity. Moving from physical simulators into digital simulators based upon VR technology thus challenges us as HCI researchers to figure out how to design social fidelity into immersive training simulators. We explore social fidelity theoretically and technically by combining core conceptual work from CSCW research to the design experimentation of social fidelity for maritime safety training. We argue that designing for social fidelity in VR simulators requires designers to contextualize the VR experience in location, artifacts, and actors structured through dependencies in work allowing trainees to perform situational awareness, coordination, and communication which are all features of social fidelity. Further, we identify the risk of breaking the social fidelity immersion related to the intent and social state of the participants entering the simulation. Finally, we suggest that future designs of social fidelity should consider not only trainees in the design, but also the social relations created by the instructors’ guidance as part of the social fidelity immersion.
Recent advances, especially in deep learning, allow to effectively detect ship targets in surveillance videos. However, the translation of these detections to the real-world locations of ships has not been sufficiently explored. The common approach in the literature is using a transformation matrix to convert a pixel to a real-world coordinate. However, this approach has three shortcomings: first, a set of reference point pairs has to be manually prepared to establish the matrix; second, the matrix always maps a pixel to the same real-world coordinate, ignoring that there is no one-to-one correspondence between discrete pixel coordinates and continuous real-world coordinates; third, this approach can only work with one camera. In light of this, we propose a technique PixelToRegion that explicitly takes into account the uncertainty in coordinate conversion by mapping each pixel to a spatial polygon. Next, we propose a new algorithm MCbSLE that can estimate ship locations using pixel sets from multiple cameras. The precision of location estimation by MCbSLE is enhanced through spatial intersection between polygons from different cameras. Experiments are conducted under 16 carefully designed multi-camera settings to evaluate MCbSLE wrt four factors: different ports, the number of cameras, the distance between cameras, and camera headings. Results on one-day ship trajectory data show that (1) an 79.8% accuracy in the number of coordinates can be achieved by MCbSLE when there are no more than 10 ships in camera views; (2) using multiple cameras can improve the precision of location estimation by one order of magnitude compared with using one camera.