Towards mixed reality applications on light-field displays

Mixed reality applications have been traditionally developed for stereoscopic or auto-stereoscopic displays scaling from AR glasses to Powerwalls and CAVEs. These approaches however, all suffer from the inherent shortcomings of the underlying technology such as limitations of stereoscopic rendering, tracking latency, low resolution and potentially uncomfortable wearable components. In this paper we present a novel rendering method that overcomes these limitations by augmenting the virtual images rendered on light-field displays. Our ultimate aim is to blur the line between real and virtual as much as possible and have a gadgetless, completely believable experience where the ambient environment is able to seamlessly augment the rendered image on the screen. Our approach makes it possible for the ambient environment to augment the virtual content rendered on the light-field display in real-time and in a view-dependent way which was not possible with earlier techniques.