Speaker: Prof. Josef Kittler. University of Surrey, UK.
Title: 3D morphable face model and its applications
Abstract: 3D Morphable Face Models (3DMM) have been used in face recognition for some time now. They can be applied in their own right as a basis for 3D face recognition and analysis involving 3D face data. However their prevalent use over the last decade has been as a versatile tool in 2D face recognition to normalise pose, illumination and expression of 2D face images. It has the generative capacity to augment the training and test databases for various 2D face processing related tasks. It can expand the gallery set for pose invariant face matching. For any 2D face image it can furnish complementary information, in terms of its 3D face shape and texture. It can also aid multiple frame fusion by providing the means of registering a set of 2D images. A key enabling technology for this versatility is 3D face model to 2D face image fitting. The recent developments in 3D model to 2D image fitting will be discussed. They include the use of symmetry to improve the accuracy of illumination estimation, multistage close form fitting to accelerate the fitting process, modifying the imaging model to cope with 2D images of low resolution, and building albedo 3DMM. These various enhancements will be overviewed and their merit demonstrated on a number of face analysis related problems.
Brief Biography: Josef Kittler is professor of Machine Intelligence at the Centre for Vision, Speech and Signal Processing, University of Surrey. He received his BA, PhD and DSc degrees from the University of Cambridge. He teaches and conducts research in the subject area of Signal Processing and Machine Intelligence, with a focus on face biometrics, and anomaly detection. He published a Prentice Hall textbook on Pattern Recognition: A Statistical Approach and several edited volumes, as well as more than 700 scientific papers, including in excess of 180 journal papers. He serves on the Editorial Board of several scientific journals in Pattern Recognition and Computer Vision. He became Series Editor of Springer Lecture Notes on Computer Science in 2004. He served as President of the International Association for Pattern Recognition 1994-1996. He was elected Fellow of the Royal Academy of Engineering in 2000. In 2006 he was awarded the KS Fu Prize from the International Association for Pattern Recognition, for outstanding contributions to pattern recognition. In 2008 he was awarded the IET Faraday Medal and in 2009 he became EURASIP Fellow.
Speaker: Prof. Michael Bronstein. University of Lugano, Switzerland / Intel Perceptual Computing, Israel
Title: Geometric deep learning
Abstract: The past decade in computer vision research has witnessed the re-emergence of “deep learning” and in particular, convolutional neural network techniques, allowing to learn task-specific features from examples and achieving a breakthrough in performance in a wide range of applications. However, in the geometry processing and computer graphics communities, these methods are practically unknown. One of the reasons stems from the facts that 3D shapes (typically modeled as Riemannian manifolds) are not shift-invariant spaces, hence the very notion of convolution is rather elusive. In this talk, I will show some recent works from our group trying to bridge this gap. Specifically, I will show the construction of intrinsic convolutional neural networks on meshes and point clouds, with applications such as finding dense correspondence between deformable shapes and shape retrieval.
Brief Biography: Michael Bronstein is a professor in the Faculty of Informatics at the University of Lugano (USI), Switzerland and a Research Scientist at the Perceptual Computing group, Intel, Israel. Michael got his B.Sc. in Electrical Engineering (2002) and Ph.D. in Computer Science (2007), both from the Technion, Israel. His main research interests are theoretical and computational methods in spectral and metric geometry and their application to problems in computer vision, pattern recognition, computer graphics, image processing, and machine learning. His research appeared in international media and was recognized by numerous awards. In 2012, Michael received the highly competitive ERC starting grant. In 2014, he was invited as a Young Scientist to the World Economic Forum, an honor bestowed on forty world’s leading scientists under the age of 40. Besides academic work, Michael is actively involved in the industry. He was the co-founder of the Silicon Valley start-up company Novafora, where he served as VP of technology (2006-2009), responsible for the development of algorithms for large-scale video analysis. He was one of the principal inventors and technologists at Invision, an Israeli startup developing 3D sensing technology acquired by Intel in 2012 and released under the RealSense brand. This technology can now be found in new generation computers from all the major brands.
Speaker: Prof. Miguel Chover. University of Jaume I, Spain
Title: Democratizing Game Development
Abstract: Videogames creation is a really complex process where the participation of a multidisciplinary team, as well as the use of tools to assist the creation of content, is required. In the mid-1990s, game engines appear with the intention to reuse the software and facilitate the creation of new games. The value of this separation introduced between the game and the game engine is essential in today’s industry. Thus, it is easy to create new games, simply making art, levels, weapons, characters, vehicles and behaviour without changing the “engine”. This kind of software has evolved in recent years including editors that can visually build the interaction and the behaviours of the characters and objects. Most of these applications use the concept of game object for scene creation, where different types of components and properties to extend their functionality are added, including scripting. In recent years, the increasing popularity of casual games for mobile and web has promoted the development of new game editors for creating 2D games where the behaviour of the game elements is visually described and it is not necessary to have programming skills.
At this point, we can say that, the creation of videogames has been democratized, so that anyone interested, without advanced programming skills, can create games for devices such as mobile phones or game consoles. In this presentation, the main characteristics of this new paradigm of content creation are analysed. Moreover, it will be explained how to make arcade games without using computer knowledge and complex data structures. Finally, we will propose some ideas to describe interactive scenes that can be applied for behaviour specification of game objects in the new game editors.
Brief Biography: Miguel Chover is full professor in the Department of Computer Languages and Systems at the University Jaume I of Castellón. He received his PhD in computer science from the Polytechnic University of Valencia in 1996. He is currently director of the Degree in Design and Game Development, director of the Centre for Interactive Visualization and member of the Institute for New Imaging Technologies of the University Jaume I. His research interests include geometric modelling, real-time rendering and computer game technology. He is vice president and member of the Spanish Association of Computer Graphics (EUROGRAPHICS S.E.), member of the General Council of Scientific Computer Society of Spain (SCIE) and member of the executive committee of the Spanish Society for Videogame Science (SECiVi). He has participated as a founding partner in the Spin-off Paidia Technologies and in the technology-based company Gamesonomy.
Speaker: Prof. Manuel Ujaldón, CUDA Fellow @ Nvidia Corporation. Associate Professor Univ. of Malaga (Spain).
Title Tutorial: Programming GPUs with CUDA
Abstract: Over the past decade, GPU Computing has evolved from an obscure niche to a pervasive area of high performance computing in both academia and industry. Applications from hundreds of scientific domains have been ported to GPUs with substantial speedups, and several of the world’s fastest supercomputers rely on GPUs for their outstanding performance and energy efficiency. This success story has been enabled by a steady and joint development of the hardware and novel algorithmic techniques for fine-grained parallelism, along with new programming environments, libraries, toolchains, accessible programming interfaces and industry-standard languages such as C. GPU Computing today is easier and more accessible than ever, but at the same time, the challenge remains to achieve substantial levels of performance.
This tutorial provides the basic pillars of CUDA architecture and its programming, with a final emphasis on deep learning to figure out how to exploit GPU capabilities from existing tools, libraries and middleware in that increasingly popular area.
Brief Biography: Manuel Ujaldon is Prof. of Computer Architecture at the Univ. of Malaga (Spain) and CUDA Fellow at Nvidia. He worked in the 90’s on parallelizing compilers, finishing his PhD in 1996 by developing a data-parallel compiler for sparse matrix and irregular applications. Over this period, he was part of the HPF and MPI Forums, working as post-doc in the CS Dept. of the Univ. of Maryland (USA). Last decade he started working on the GPGPU movement early in 2003 using Cg, and wrote the first book in spanish about programming GPUs for general purpose computing. He adopted CUDA when it was first released, then focusing on image processing and biomedical applications. Over the past five years, he has published more than 50 papers in journals and international conferences in these two areas. Dr. Ujaldon has been awarded as NVIDIA Academic Partnership 2008-2011, NVIDIA Teaching Center since 2011, NVIDIA Research Center since 2012, and finally CUDA Fellow. Over the past four years, he has taught around 60 courses on CUDA programming worldwide sponsored by Nvidia, including keynotes and tutorials in ACM/IEEE conferences and academic programs in universities of 18 different countries.