Registration includes lunch, Coffee Break and access to all keynote talks, poster session and master thesis presentations.
14.00 – 15.00 Lunch + Poster Session
15.00 – 15.45 Keynote talk. ‘Vision challenges in the project CLOTHILDE: Cloth manipulation learning from demonstration‘ by Dr. Guillem Alenyà (Former Head of the robotic Department at IRI)
Abstract: The CLOTHILDE project’s aim is to establish the foundations of versatile cloth manipulation by robots. The theoretical ground will be to link machine learning and computational topology methods, so as to come up with a theory of cloth deformation under manipulation leading to a general framework for robots to learn to manipulate garments from human demonstrations. Perception is one of the three pillars of the project and plays a fundamental role in understanding the human and the garment state and how it evolves after manipulations. In particular, perception of clothing is challenging because the variety of possible deformations and strong self occlusions. We will show some examples that demonstrate the challenges and introduce possible solutions.
Bio: Guillem Alenyà holds a tenured position at Spanish Scientific Research Council CSIC. He received a PhD degree (Doctor Europeus) from UPC in 2007 with a work on mobile robot navigation using active contours, which he partly developed at the Robosoft company in France, where he was supported by a EU-FP6 Marie-Curie scholarship. He has been visitor at KIT-Karlsruhe (2007), INRIA-Grenoble (2008) and BRL-Bristol(2016). He has participated and coordinated numerous scientific and technological transfer projects involving image understanding, next-best-view, rule learning and plan execution tasks. From Jan 2009 to Apr 2011 he served as head of the Robotics Department and is currently part of IRI’s board.
15.45 – 16.15 Coffee Break
Abstract: In this talk, I will present the evolution of my research during the last ten years, starting from my PhD at CVC to my postdocs until my current position as associate professor at ETS Montreal. I will introduce two topics that accompanied me throughout my research: Efficient Computation and Learning with reduced supervision. I will show how these two topics have evolved in the field of computer vision and machine learning and, in particular, in my research. Finally, I will present my current research and the challenges that I foresee for the next ten years in the field.
Bio: Marco Pedersoli obtained his PhD in computer science in 2012 at the Autonomous University of Barcelona and the Computer Vision Center. He was research fellow in computer vision and machine learning at KU Leuven with professor Tinne Tuytelaars and later at INRIA Grenoble with Doctors Jakob Verbeek and Cordelia Schmid. Since 2017, he is associate professor at ETS Montreal, where he is part of LIVIA, the laboratory of vision and artificial intelligence. He has authored more than 30 articles in international conferences and computer vision journals.
16.45 – Student Awards presentation