Image processing techniques for mixed reality and biometry

This thesis work is focused on two applicative fields of image processing research, which, for different reasons, have become particularly active in the last decade: Mixed Reality and Biometry. Though the image processing techniques involved in these two research areas are often different, they share the key objective of recognizing salient features typically captured through imaging devices. Enabling technologies for augmented/mixed reality have been improved and refined throughout the last years and more recently they seems to have finally passed the demo stage to becoming ready for practical industrial and commercial applications. To this regard, a crucial role will likely be played by the new generation of smartphones and tablets, equipped with an arsenal of sensors connections and enough processing power for becoming the most portable and affordable AR platform ever. Within this context, techniques like gesture recognition by means of simple, light and robust capturing hardware and advanced computer vision techniques may play an important role in providing a natural and robust way to control software applications and to enhance onthe- field operational capabilities. The research described in this thesis is targeted toward advanced visualization and interaction strategies aimed to improve the operative range and robustness of mixed reality applications, particularly for demanding industrial environments... [edited by Author]