(6 January 2016) The Marr prize, awarded every two years to the best scientific article on artificial vision, in its latest edition went to the Bruno Kessler Foundation in Trento for a study conducted in collaboration with Microsoft and Carnegie Mellon University. The work was considered the best among the 1,700 submitted projects during the most important international conference in the industry, the “IEEE International Conference on Computer Vision”, held last December in Santiago de Chile. The study was carried out by Samuel Rota Bulo, researcher of the TeV (Technologies of Vision) Unit of the ICT Centre of the Bruno Kessler Foundation, directed by Paolo Traverso. With this innovative work, called “Deep Neural Decision Forests”, researchers were able to unify the two techniques used in the field of automatic classification of images (“deep network” and “random forest”) and develop a new system (that held the best performance on databases used in the scientific community). The TeV research unit, currently led by Stefano Messelodi, is no stranger to international first class results. A study conducted by researchers Oswald Lanz and Elisa Ricci in collaboration with the University of Trento crushed the international competition during Acmmm 2015, the most important Multimedia world conference, held last October in Brisbane, Australia. In this case, the authors presented a method to interpret more accurately the behavior and interaction between people in a social setting. The scientific study proposed a system to efficiently integrate the information extracted from a video of a scene with that sent from sensors worn by people in the scene. The TeV Unit also received European funding for “Replicate”, a project in the field of augmented reality that will kick off in 2016 and which will be coordinated by the Bruno Kessler Foundation. (Red)
