Caruso: augmenting users with a tenor's voice

Abstract

In this poster paper we describe an ongoing project that aims at providing users with the experience to sing like a tenor from the 20th century. We combine 3D body tracking, face recognition and morphing, sound synthesis and 3D character rendering into an interactive media application. Although this project is still in a preliminary stage, first prototypical results are encouraging to continue this work towards a media installation for a science fair.