Infotech Oulu Lecture Series
Robust methods for spatial and temporal alignment
Lecturer: Dr. Stefanos Zafeiriou, Imperial College London, UK
Date: Tuesday, June 3, 2014
The past decade computer vision (CV), the area of computer science that specialises in automatic analysis and understanding of visual data, has witnessed tremendous developments due to the abundance of visual data, the development of powerful features (such as SIFT, HoGs, LBPs etc.) and the prevalence of Statistical Machine Learning (SML) , which promised and delivered many powerful learning methodologies. But what is then the difference between modern computer vision and statistical machine learning? The difference is alignment. In particular, in computer vision before defining SML algorithms, or even applying a simple (dis)-similarity between spatial or temporal data, we first need to remove any spatial or temporal discrepancies. In this tutorial, I will motivate and describe recent methodologies of my group regarding spatial alignment of facial images, as well as, temporal alignment of facial behaviour.
Dr. Stefanos Zafeiriou, since 2012, is a Lecturer (equivalent to Assistant Professor) in Computer Vision and Pattern Recognition in the Department of Computing, Imperial College London. He has been awarded one of the prestigious Junior Research Fellowships (JRF) from Imperial College London in 2011. He received the B.Sc, degree in Informatics with highest honors and the Ph.D degree in Informatics in 2003 and 2007, respectively, from the Aristotle University of Thessaloniki, Greece. He has been conducting research in automatic facial analysis for more than 9 years participating in more than 10 EU and British research projects. From 2003 to 2007 he was a research and teaching assistant in the Department of Informatics at Aristotle University of Thessaloniki, Greece. During 2007-2008 he was a Research Associate in the same Department. During 2008-2011 he was a Research Associate/Research Fellow in the Departments of Electrical and Electronic Engineering and Computing, Imperial College London, UK. In 2009 he was a Visiting Researcher at the University of Houston, Texas, USA. He has received various scholarships and awards during his undergraduate, Ph.D. and postdoctoral studies. Dr. Zafeiriou is an associate editor of Image and Vision Computing Journal and of IEEE Transactions on Cybernetics (previously IEEE Transactions on Systems, Man, and Cybernetics Part B). He has co-authored more than 100 technical papers including more than 30 journal papers in face recognition, face tracking/alignment, modelling of facial motion and deformation and novel machine learning approaches to modelling facial behaviour, as well as more than 20 papers in top vision and learning conferences such as CVPR, ICCV, ECCV, ICML, ECML etc.
More information: Guoying Zhao
Last updated: 19.5.2014