We present a novel interaction method for augmented industrial maintenance based on a "magic mirror" interface and virtual motion buttons. The system includes a video camera for object tracking, a video\depth camera for capturing user gestures, a projector for displaying technical instruction to the operator and a LCD monitor providing feedback of the virtual buttons. The operator can trigger maintenance commands by directional swift of the hands in regions sensitive to motion speed and direction. The main advantage of the presented interface is that it can work in realistic industrial conditions: (i) operators wearing gloves, (ii) operators handling tools, (iii) presence of moving machinery and personnel in the background. We measured the performances of the system with a laboratory test and we proved the feasibility with an automotive inspection test case. We calculated an average interaction time below 2 seconds and an error rate lower than 5%. However, we found some performances limitations if the operator is handling tools.
Magic mirror interface for augmented reality maintenance: An automotive case study / Fiorentino, Michele; Radkowski, Rafael; Boccaccio, Antonio; Uva, Antonio Emmanuele. - (2016), pp. 160-167. (Intervento presentato al convegno 13th ACM International Working Conference on Advanced Visual Interfaces, AVI 2016 tenutosi a Bari, Italy nel June 7-10, 2016) [10.1145/2909132.2909268].
Magic mirror interface for augmented reality maintenance: An automotive case study
FIORENTINO, Michele;BOCCACCIO, Antonio;UVA, Antonio Emmanuele
2016-01-01
Abstract
We present a novel interaction method for augmented industrial maintenance based on a "magic mirror" interface and virtual motion buttons. The system includes a video camera for object tracking, a video\depth camera for capturing user gestures, a projector for displaying technical instruction to the operator and a LCD monitor providing feedback of the virtual buttons. The operator can trigger maintenance commands by directional swift of the hands in regions sensitive to motion speed and direction. The main advantage of the presented interface is that it can work in realistic industrial conditions: (i) operators wearing gloves, (ii) operators handling tools, (iii) presence of moving machinery and personnel in the background. We measured the performances of the system with a laboratory test and we proved the feasibility with an automotive inspection test case. We calculated an average interaction time below 2 seconds and an error rate lower than 5%. However, we found some performances limitations if the operator is handling tools.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
