Enhancing Sight with Machine Learning and Augmented Reality
"Becoming blind" is commonly ranked in the top three fears that people have (the others are paralysis and cancer), and it's no wonder - our world is predominantly visual and blindness robs people of a great deal of independence. 40 million people are living with legal blindness which often prevents them from seeing the face of loved ones, reading to themselves, driving, and walking in crowded or dimly lit spaces. We have developed a sophisticated array of software to detect a wide range of objects and scenarios which we have built into a pair of almost regular looking glasses. Our aim is to boost any residual vision that a person might have to the point in which they can use their own memory of vision to see and function more effectively in everyday life. Machine learning and Computer Vision are at the heart of our software as they offer a wide range of abilities: from detecting objects and scenes to learning and tracking any arbitrary object, face or person.
Luca has obtained a joint MSc in Computer Engineering between the Polytechnic University of Turin and Telecom Paris Tech. At the moment he is at the third year of his PhD program within the Torr Vision group at the University of Oxford. The focus of his doctorate is learning representations from video when very little supervision is present - the so called one-shot learning scenario. He is interested in applying these techniques to the problem of arbitrary object tracking, which is a key component of many AI-equipped video processing systems.