Multimodal cognitive assistance for EMS using augmented reality at the edge

The University of Virginia developed an AR cognitive assistant system that supports first responders through real-time inference of incident context based on observations at the scene and provides them with context-dependent feedback on the timely and safe execution of response actions. This session will detail their work with first responders to create a multimodal dataset, EgoExoEMS, that combines different views of EMS procedures in emergency scenarios.