An Introduction to the 3rd Workshop on Egocentric (First-Person) Vision

Egocentric vision provides a unique perspective of the visual world that is inherently human-centric. Since egocentric cameras are mounted on the user (typically on the user´s head), they are naturally primed to gather visual information from our everyday interactions, and can even act on that information in real-time (e.g. for a vision aid). We believe that this human-centric characteristic of egocentric vision can have a large impact on the way we approach central computer vision tasks such as visual detection, recognition, prediction, and socio-behavioral analysis. By taking advantage of the first-person point-of-view paradigm, there have been recent advances in areas such as personalized video summarization, understanding concepts of social saliency, activity analysis with inside-out cameras (a camera to capture eye gaze and an outward-looking camera), recognizing human interactions and modeling focus of attention. However, in many ways people are only beginning to understand the full potential (and limitations) of the first-person paradigm. In the 3rd workshop on Egocentric (First-Person) Vision, we bring together researchers to discuss emerging topics such as: Personalization of visual analysis; Socio-behavioral modeling; Understanding group dynamics and interactions; Egocentric video as big data; First-person vision for robotics; and Egographical User Interfaces (EUIs).