Golem Workshop. 1st July 2014 (Universidad de Zaragoza / I3A)
Henrik Wann Jensen: Crossing the Uncanny Valley
Marc Swerts seminar talk.
The Golem consortium is proud to announce the organization of a multidisciplinary Workshop co-organized with the CEIG conference and to be held on the 1st of July 2014 (Universidad de Zaragoza / I3A). This multidisciplinary workshop is intended to bring together Ph.D. students in the various fields related with Virtual Realistic Humans like Computer Graphics, Computer Animation or Speech, and discussed above the aforementioned fields to discuss their research in an international forum. The workshop will provide students with an opportunity to:
- Present their research work in a relaxed and supportive environment;
- Receive constructive feedback and suggestions from peers and experienced faculty;
- Gain an overview of the breadth and depth of biomedical engineering systems and technologies;
- Obtain insight into directions for their research taken by other doctoral candidates;
- Discuss concerns about research, supervision, the job market, and other issues;
- Network with peers and future colleagues.
This Workshop is co-organized with Microsoft that will donate a prize for the best project. Please download the call for abstracts of the Workshop and consult the basis for the competition here.
News: Notification of the list of presenters at the Golem Workshop. Next find the list of accepted speakers and their presentations:
|C. Aliaga: “Biophysically-based Aging of Human Skin Appearance”.|
|J. I. Echevarria: “Hair Stylization”.|
|R. Esporrin: “Bidirectional Clustering in Point Based Global Illumination”.|
|F. Gaspar: “Good RGB-D points for Pose Estimation and 3D Registration”.|
|A. Jarabo: “Separable Subsurface Scattering”.|
|S. Azevedo: “Perceptually-based enhancements on advanced computational displays”.|
|D. Castán: “Advances on Audio Segmentation and Audio Content Description for Multimedia Documents”.|
|D. Figueira: “Human activity recognition and prediction”.|
|J. Llombart: “Study of the impact of articulatory features in total variability space on hybrid acoustic models and speaker characterization”.|
|D. Martínez: “Dysarthria Intelligibility Assessment”.|
|J. Olcoz: “Study of unsupervised learning techniques in automatic speech recognition system for unrestricted domains”.|
In the end, the Workshop Advisory Panel decided to award on the 1st July in Zaragoza Jose I. Echevarria and his presentation “Hair Stylization” with the 1st position award of the Golem Workshop. The prize included a Microsoft Kinect sensor kindly donated by Microsoft Portugal.
Furthermore, we have the pleasure to count with two outstanding invited speakers and members of the Workshop Advisory Panel:
- Prof. Hernik Wann Jensen
- Prof. M.G.J. Swerts
Short Bio: Henrik Wann Jensen is a professor at the University of California at San Diego, where he is working in the computer graphics lab. His research is focused on realistic image synthesis, global illumination, rendering of natural phenomena, and appearance modeling. His contributions to computer graphics include the photon mapping algorithm for global illumination, and the first technique for efficiently simulating subsurface scattering in translucent materials. He is the author of “Realistic Image Synthesis using Photon Mapping,” AK Peters 2001. He has rendered images that have appeared on the frontcovers of the National Geographic Magazine and the SIGGRAPH proceedings. He previously worked at Stanford University, Massachusetts Institute of Technology (MIT), Weta, Pixar, and at mental images. He received his M.Sc. and Ph.D. in Computer Science from the Technical University of Denmark. He is the recipient of an Academy Award (Technical Achievement Award) from the Academy of Motion Picture Arts and Sciences for pioneering research in rendering translucent materials. He also received a Sloan Fellowship, and was selected as one of the top 10 scientists by Popular Science magazine.
Seminar Talk: Crossing the Uncanny Valley
Rendering realistic human beings is one of the great challenges in computer graphics. It involves simulating the appearance of human skin and hair in addition to capturing all the geometric detail. In this talk I will discuss our work on rendering human skin and hair. For human skin it is necessary to capture the light scattering by the different layers of the skin and take into account each layers unique properties. This requires a full simulation of subsurface scattering to capture the translucent appearance of skin. To render human hair it is necessary to include the microstructure of the hair – the cuticle scales – to properly capture how light is scattered by a single human hair.
Short Bio: Marc Swerts is a full professor in the Department of Communication and Information Science of the Faculty of Humanities at Tilburg University (The Netherlands). His current research aims to get a better understanding of how speakers exploit verbal of non-verbal forms of communication to exchange information with their addressees. He has served on the editorial board of three major journals in the field of language and speech research, and has been the current co-editor-in-chief of Speech Communication. He was also elected as one of the two distinguished lecturers of the International Speech Communication Association (ISCA) for 2007-2008.
Seminar Talk: Nonverbal characteristics of children in game-based interactions
Children make extensive use of facial expressions when interacting with others. One typical characteristic is that these expressions tend to reveal whether children feel positive or negative about a specific event. In this talk, I will present results on contextual, developmental, and cultural factors that can determine the extent to which children display affect. These results are primarily based on studies of game-based interactions, where specific techniques were used that ensure that the analysed data were ecologically valid. These games were played by children (different age groups, different cultures) in different contexts: that is while playing alone, while playing with other children, or while playing with a robot. This talk is based on joint work with Suleman Shahid and Emiel Krahmer.