IMI Distinguished Lecture

IMI Virtual Reality Theatre: 50 Nanyang Drive, Research Techno Plaza, XFrontiers, Level 02-01
Singapore 637553

On Friday June 11th at 2pm, Professor Enhua Wu of Academica Sinica, Beijing and the University of Macau in China will give a distinguished lecture at the IMI Virtual Reality Theatre.

His talk is entitled "Time-Varying Simulation to Natural Objects".


 

Invited Speakers for CGI'10

"Real Time Illumination Estimation of Outdoor Scenes Based on Images for Augmented Reality"

Qunsheng Peng

State Key Lab of CAD&CG, Zhejiang University, China

Chair:

Enhua Wu

Abstract

Illumination consistency plays an important role for augmented reality. However, acquisition of the light sources of the real environment is usually difficult. In this talk, we present a set of novel methods for evaluating the light conditions of outdoor scenes based on sample images. We focus our attention on estimating the sunlight and skylight of static outdoor scenes for online video captured at a fixed viewpoint with panning.

Under the assumption that sunlight is directional and skylight is ambient, we derive an analytic expression relating the parameters of the sunlight and skylight of a scene with the image statistics. Based on this model, a framework for real-time estimation of outdoor illumination is established. We then extend the skylight to a uniformly distributed area light source. By decomposing each sample image of a fixed outdoor scene into a linear combination of a sunlight base image and a skylight base image, we develop a robust approach for estimating outdoor illumination. Our approach is capable of handling occasional motions in outdoor scenes, such as swaying trees and intrusive pedestrians.

We further present a novel approach for outdoor illumination estimation without off-line learning. This method involves no reflectance assumption and permits objects with complex reflectance such as specular, or anisotropic surfaces hence suitable for online video processing such as augmented reality, shadow detection, relighting, color constancy and illumination normalization. None of the above approaches requests information of scene geometry, thus getting rid of the difficulties in reconstructing the 3D geometry of large-scale outdoor scenes. As the illumination consistency is ensured, virtual objects can be seamlessly integrated into a live video.

About the Speaker

Prof. Peng graduated from Beijing Mechanical College in 1970 and received a Ph.D from the Department of Computing Studies, University of East Anglia, UK in 1983. He is currently a professor in State Key Lab of CAD&CG, Zhejiang University and serves as the vice chairman of the Lab Academic Committee. His research interests include realistic image synthesis, virtual reality, bio-molecule graphics and scientific visualization. In these fields, he has authored and coauthored more than two hundred journal and conference papers, including ACM SIGGRAPH, ACM VRST, ICCV, Eurographics, Pacific Graphics, etc. He won the best paper award of Eurographics’89 and received Computer Graphics Achievements Award of China at Chinagraph’2000. He is a member of the editorial board of The Visual Computer, Journal of Computer Science and Technology and several Chinese journals.

 

 

"Shape and Image Cognition, Construction and Compression via Tools from Differential Geometry"

Franz-Erich Wolter

Welfenlab, Institute of Man-Machine-Communication, Leibniz University Hannover, Germany
[Download Presentation Slides]

Chair:

Nadia Magnenat-Thalmann

Abstract

We shall describe how concepts from differential geometry especially from Riemannian geometry have been providing powerful tools creating major advances in geometric modeling, geometry processing and image analysis dealing with the topics presented in the title of this address. This talk includes a retrospective compiling contributions of the author`s works showing how concepts from local and global differential geometry have introduced new methods into geometric modeling and shape interrogation and classification finally ending with modern state of the art research on geometry processing and image processing. We start sketching early contributions on curvature analysis of surface constructions and surface intersections. Next we focus on cut loci, the medial axis and its inverse in Euclidean and Riemannian worlds. This work starts with basic medial axis results presented by the the author in the early nineties when the medial axis was not considered as a main stream subject in the respective scientific communities. Those results state: The Medial Axis Transform can be used to reconstruct, modify and design a given shape ("Shape Reconstruction Theorem"). Under some weak assumptions the medial axis contains the essence of the topological shape of the geometric object as it is a deformation retract of the given shape ("Topological Shape Theorem"). Therefore the medial axis contains the homotopy type of the given shape. We present recent results showing how geodesic Voronoi diagrams, geodesic medial axis and its inverse can be computed in 3d- or higher dimensional Riemannian spaces. The "medial axis inverse" allows to construct a medial modeller providing efficient features for shape optimization with respect to shape dependent mechanical properties. A major part of this seminar starting with works in the nineties at the author`s lab is dedicated to discussing how "efficient finger prints" useful for indexing and clustering digital data collections can be derived from spectra of Laplace operators being naturally associated with geometric objects such as surfaces and solids as well as (coloured) images including medical 2d- and 3d-images. Recently the latter works obtained particular attention in the area of medical imaging.

About the Speaker

Dr. F.-E. Wolter has been a full professor of computer science at the Leibniz Universität Hannover (LUH) since the winter term of 1994/1995, where he heads the Institute of Man-Machine-Communication and directs the Division of Computer Graphics and Geometric Modeling called Welfenlab. Before coming to Hannover, Dr. Wolter held faculty positions at the University of Hamburg (in 1994), MIT (1989-1993) and Purdue University (1987-1989). Prior to this he developed industrial expertise as a software and development engineer with AEG in Germany (1986-1987). Dr. Wolter obtained his Ph.D. in 1985 from the department of mathematics at the Technical University of Berlin in the area of Riemannian manifolds. In 1980 he graduated in mathematics and theoretical physics from the Free University of Berlin. At MIT Dr. Wolter codeveloped the geometric modeling system Praxiteles for the US Navy. Since then he has been publishing various papers that broke new ground applying concepts from differential geometry and topology on problems and design of new methods used in geometric modeling and CAD systems as well as shape and image analysis. These works include pioneering contributions on medial axis theory, the computation of medial axes and Voronoi diagrams in Riemannian space as well as pioneering works on Laplace spectra as finger prints for multi dimensional geometric objects and images. During the last years research on VR-systems with an emphasis on haptic and tactile perception has been subject of Dr. Wolter`s research in Hannover. More recently research at the Welfenlab includes the development of medical imaging systems and bio-mechanical simulation systems. Dr. Wolter is a research affiliate of MIT.

 

 

"Emotion-based Personalised Digital Media Experience"

Olga Sourina

Division of Information Engineering, School of Electrical & Electronic Engineering, Nanyang Technological University, Singapore

Chair:

Daniel Thalmann

Abstract

New forms of human-centric and human-driven interaction with digital media have the potential of revolutionising entertainment, learning, and many other areas of life. The user of interactive digital media could have a new emotion-based personalized digital experience by using electro-encephalograph cap devices. The user could derive/add the content into Co-Space based on the user’s engagement/emotion.  We discuss emotion/engagement driven human interaction with digital media. The experiments on evoking emotions by music stimuli are described. Models and algorithms of quantification of brain responses to external stimuli are discussed.  In the nearest future, new affordable electro-encephalograph cap devices with wireless data transmission would bring the emotion-based personalized digital experience to any user’s location making such interfaces more mobile.

About the Speaker

Dr. Olga Sourina received her MSc in Computer Engineering from Moscow Engineering Physics Institute (MEPhI) in 1983, and her PhD in Computer Science from NTU in 1998. Dr Sourina worked as a software engineer, then as a Research Scientist in MEPhI. For her scientific achievements Dr. Sourina was awarded the honorary diploma of the Academy of Sciences of USSR, the Silver Medal of the National Exhibition Centre of USSR, and the Medal of the Ministry of Education of USSR. After receiving her PhD from NTU she worked as a Research Fellow in the Centre for Graphics and Imaging Technology (CGIT), NTU. Then, she worked as a Senior Scientist in Institute of Computing for Physics and Technology in Russia. Since 1 December 2001 Dr Sourina has been working as an Assistant Professor in NTU.  Her research interests are in interactive digital media IDM (particularly in visual data mining, virtual reality, and visual and haptic interfaces), and Biomedical Engineering (visual analysis and quantification of brain responses, virtual surgery). She collaborates with SBS, MAE, SCE, and medical doctors from SGH (Singapore General Hospital), NUH (National University Hospital), and NNI (National Neuroscience Institute).

Dr Sourina has more than 70 publications including more than 20 research papers in international refereed journals and 3 books. She presented 30 papers at international conferences. She is a member of program committee of international conferences including conference on Cyberworlds, a member of the editorial board of International Journal “Computer Graphics & Geometry”, a member of the IEEE Computer Society and a member of Biomedical Engineering Society.

 

 

"One step at a time: animating virtual characters based on foot placement"

Arjan Egges

Games and Virtual Worlds group, Department of Information and Computing Sciences, Utrecht University, The Netherlands

Chair:

Nadia Magnenat-Thalmann

Abstract

Especially in constrained virtual environments, precise control of foot placement during character locomotion is crucial to avoid collisions and to ensure a natural locomotion. We present the step space: a novel technique for generating animations of a character walking over a set of desired foot steps in real-time. We will present different ways in which this approach can be used to drive a virtual character in real-time. We will show an efficient greedy nearest-neighbor approach and warp the resulting animation such that it adheres to both spatial and temporal constraints. Secondly, we will presnet a linearized interpolation scheme in the step space, allowing blending of foot steps based on feet positions, resulting in exaxt foot placement without having to do inverse kinematics.

About the Speaker

Arjan Egges is an Assistant Professor at the Games and Virtual Worlds group in the Department of Information and Computing Sciences, Utrecht University in the Netherlands. He obtained his PhD at MIRALab - University of Geneva, Switzerland on the topic of emotion and personality models, in combination with automatically generated face and body motions using motion capture data. His current research focuses on the integration of motion capture animation with navigation and object manipulation tasks, as a part of the Dutch funded GATE project.
Furthermore, he heads the motion capture lab and he teaches several courses related to games and computer animation. Arjan is also an associate editor of the Computer Animation and Virtual Worlds journal published by Wiley and he is one of the co-founders of the annual Motion in Games meeting.