Representing indoor location of objects on wearable computers with head-mounted displays

With head-mounted displays becoming more ubiquitous, the vision of extending human object search capabilities using a wearable system becomes feasible. Wearable cameras can recognize known objects and store their indoor location. But how can the location of objects be represented on a wearable device like Google Glass and how can the user be navigated towards the object? We implemented a prototype on a wearable computer with a head-mounted display and compared a last seen image representation against a map representation of the location. We found a significant interaction effect favoring the last seen image with harder hidden objects. Additionally, all objective and subjective measures generally favor the last seen image. Results suggest that a map representation is more helpful for gross navigation and an image representation is more supportive for fine navigation.