International Conference on
Computer Animation and Social Agents
October 17 - 19, 2005
CASA 2005
Organized by: The Hong Kong Polytechnic University, City University of Hong Kong, The Computer Graphics Society (CGS)
Sponsored by: K.C. Wong Education Foundation
Keynote 1: Animation Reuse via Geometry Processing
Craig Gotsman
Technion - Israel Institute of Technology
Abstract
Generation compelling 3D animation sequences is a time- and money-consuming endeavor. Thus, the ability to reuse existing
animation data in order to generate new animation sequences has major practical importance. This talk will survey methods to
reuse animation data by applying tools originally developed for static geometry and mesh processing.
Keynote 2: Interfaces for Controlling Human Characters
Jessica Hodgins
School of Computer Science, Carnegie Mellon University, USA
Abstract
Computer animations and virtual environments both require a controllable source of motion for their characters. Most of the
currently available technologies require significant training and are not useful tools for naive users. Over the past few years, we
have explored several different solutions to this problem. Each solution relies on the information about natural human motion
inherent in a motion capture database. For example, the user can sketch an approximate path for an animated character which
is then refined by searching a motion graph. We can also find a natural looking motion for a particular behavior based on sparse
constraints from the user (foot contact locations and timing, for example) by optimizing in a low-dimensional, behavior-specific
space found from motion capture. Motion capture data can also be used to bias planning algorithms toward more natural postures
for constrained manipulation tasks such as placing a large box into a car trunk. And finally, we have developed a performance
animation system that uses video input of the user to control a swing dancing couple. The system finds the pose of the user with
machine learning techniques trained using motion capture data.
Keynote 3: Mesh Deformation Using Differential Operators
Baining Guo
Senior Researcher and Research Manager, Graphics Group, Microsoft Research Asia
Abstract
In this talk I’ll introduce a novel approach for mesh deformation using differential operators. The most distinctive feature of these
operators is that they modify the original mesh geometry implicitly through gradient field manipulation. This approach can produce
visually pleasing results for a wide variety of deformations including large deformations found with characters performing non-rigid
and highly exaggerated movements. I’ll discuss the origin of differential operators, deformation operators based on the Poisson
equation, and a recently proposed operator called volumetric graph Laplacian. I will also explain how to use these operators to
transfer non-rigid and exaggerated deformations of 2D cartoon characters to 3D meshes.
Invited Talk: Orientations and Rotations: A Coordinate-Free Pespective
Jehee Lee
Seoul National University, Korea
Abstract
There are many different ways of representing the three-dimensional orientation and rotation of a rigid object. Those includes unit
quaternions, 3x3 rotation matrices, Euler angles, rotation vectors, helical angles, and so on. Recently, it has been relatively clear
that unit quaternions and rotation matrices have advantages over the others in the sense that they parameterize three-dimensional
orientations without singularity. However, rotation vectors are still very popular because they are much easier to handle than
non-singular representations and easily converted from them using logarithmic maps. There has been a long standing argument
as to when rotation vectors can be used without worrying about singularity and when unit quaternions (or rotation matrices)
are preferable. This argument is partly due to the confusion of rotation and orientation. Rotation is circular movement and the
orientation of a rigid object refers to the state of the object being oriented. These two geometric concepts have conventionally been
considered to be the same one because the orientation of an object can be represented as a rotation from the reference orientation
given a reference coordinate system. Not surprisingly, this point of view incurred a serious confusion in programming with geometric
objects that can rotation. The goal of this talk is to provide a useful perspective of understanding three-dimensional orientations
and rotations. Our approach is inspired by affine geometry and coordinate-free geometric programming pioneered by Goldman and
DeRose. The basic idea of affine geometry is to make a distinction between points and vectors, and define operations for combining
them. We found that a similar coordinate-free geometric algebra is available between three-dimensional orientations and rotations
by making a clear distinction between them.
Day 1: Oct. 17
Keynote 1
09:00-10:00 Animation Reuse via Geometry Processing
Craig Gotsman
Session F1: Collision Detection and Deformable Objects
Chair: Shigeru Kuriyama
10:15-10:40 Multi-Resolution Collision Handling for Cloth-like Simulations
Nitin Jain, Ilknur Kabul, Naga Govindaraju, Dinesh Manocha, Ming Lin
10:40-11:05 GPU-based Intrinsic Collision Detection for Deformable Surfaces
Sai-Keung Wong, George Baciu
11:05-11:30 Implicit Midpoint Integration and Adaptive Damping for Efficient Cloth Simulation
Pascal Volino, Nadia Magnenat-Thalmann
11:30-11:55 Time-Critical Animation of Deformable Solids
Jeremie Dequidt, Damien Marchal, Laurent Grisoni
11:55-12:20 Real-time Meshless Deformation
Xiaohu Guo, Hong Qin
Session C1: Motion Synthesis
Chair: Enhua Wu
10:15-10:35 Planning Character Motions for Shadow Play Animations
Shu-Wei Hsu, Tsai-Yen Li
10:35-10:55 Rhythmic Character Animation: Interactive Chinese Lion Dance
Je-Ren Chen, Tsai-Yen Li
10:55-11:15 An interactive motion retargeting method
Li Liu, Zhao-Qi Wang, Deng-ming Zhu, Shi-Hong Xia
11:15-11:35 Inferring 3D Pose from Uncalibrated Video by Contour Matching
Xian-Jie Qiu, Zhao-Qi Wang, Shi-Hong Xia
11:35-11:55 A Deformable Model with Cellular Neural Network
Yongmin Zhong, Bijan Shirinzadeh, Gursel Alici, Julian Smith
Keynote 2
14:10-15:10 Interfaces for Controlling Human Characters
Jessica Hodgins
Session F2: Motion Capture and Retrieval
Chair: Ming Lin
12:20-12:45 From Dance Notation to Human Animation: The LabanDancer Project
Tom Calvert, Lars Wilke, Rhonda Ryman, Ilene Fox
15:25-15:50 Animating Reactive Motion Using Momentum-based Inverse Kinematics
Taku Komura, Shu Lim Ho, Rynson Lau
15:50-16:15 Entropy-Based Motion Extraction for Motion Capture Animation
Cliff So, George Baciu
16:15-16:40 Motion Control with Strokes
Masaki Oshita
16:40-17:05 Silhouette-based motion capture for interactive VR-Systems including a rear projection screen
u o
Sven Th¨ring, J¨rn Herwig, Alfred Schmitt
17:05-17:30 Automated Motion Synthesis for Dancing Characters
Gazihan Alankus, A. Alphan Bayazit, Burchan Bayazit
17:30-17:55 Motion Retrieval Based on Movement Notation Language
Tao Yu, Xiaojie Shen, Qilei Li, Weidong Geng
Session C2: Interactive Environments
Chair: Lizhuang Ma
15:00-15:20 Hierarchical kinematic synthesis of motion for avatars creation
Maria-Cruz Villa-Uriol, Falko Kuester, Nader Bagherzadeh
15:20-15:40 A New Dynamic Client-Server Scene Graph for Mobile NPRQuake on Heterogeneous Frame-
works
Gi Sook Jung, Kwon Soon Il, Park Min Woo, Won Kwang Hee, Soon Ki Jung
16:05-16:25 Adaptive Three-tiered Control Schemes for Large Multi-User Virtual Environments
Kuo-Yu Lee, Hung-Kuo Chu, Tong-Yee Lee
16:25-16:45 Programming Language Support for Collaborative Virtual Environments
Clinton Jeffery, Omar El-Khatib, Ziad Al-Sharif, Naomi Martinez
16:45-17:05 Animating Humans on Handheld Devices for Interactive Gaming
Andrew Brosnan, Simon Dobbyn, John Hamill, Carol O’Sullivan
Session PT: Poster Session
Chair: Taku Komura
17:05-17:15 Cut
Shigeru Owada, Frank Nielsen, Takeo Igarashi
17:15-17:25 Surgery Simulation with Surface Parameterization
Qiang Liu, Edmond C. Prakash
17:25-17:35 GPU-Assisted Free-Form Deformation
Takashi Kanai, Yutaka Ohtake, Kiwamu Kase
17:35-17:45 Physically Based Wire Modeling Technique for Curves and Surfaces
Jian J. Zhang, L. H. You
17:45-17:55 Towards Time-Critical Collision Detection for Deformable Objects Based on Reduced Models
Cesar Mendoza, Carol O’Sullivan
17:55-18:05 Precise Motion Capturing Using Skeletal Model Extracted from MRI
Shigeo Morishima, Shohei Nishimura, Eiji Sugisaki
Day 2: Oct. 18
Keynote 3
09:00-10:00 Mesh Deformation Using Differential Operators
Baining Guo
Session F3: Virtual Humans and Social Agents
Chair: Daniel Thalmann
10:15-10:40 Natural Head Motion Synthesis driven by Acoustic Prosody Features
Carlos Busso, Zhigang Deng, Ulrich Neumann, Shrikanth Narayanan
10:40-11:05 A Feature-Based Approach to Facial Expression Cloning
Bongcheol Park, Heejin Chung, Tomoyuki Nishita, Sung Yong Shin
11:05-11:30 Automatic Generation of Human Animation based on Motion Programming
Yueting Zhuang, Jun Xiao, Yizi Wu, Tao Yang, Fei Wu
11:30-11:55 Modeling and Simulating the Deformation of Human Skeletal Muscle Based on Anatomy and
Physiology
Robson Lemos, Jon Rokne, Gladimir V. G. Baranoski, Yasuo Kawakami, Toshiyuki Kurihara
11:55-12:20 Virtual Humans with Personalized Perception and Dynamic Levels of Knowledge
Johannes Strassner, Marion Langer
12:20-12:45 Psychological Model for Animating Crowded Pedestrians
Takeshi Sakuma, Tomohiko Mukai, Shigeru Kuriyama
Session C3: Virtual Human Modeling
Chair: Carol O’Sullivan
10:15-10:35 A Data-Driven Shape Model for Human Body Reconstruction from Photos
Hyewon Seo, Young In Yeo, Kwangyun Wohn
10:35-10:55 FacePaint: Intuitive Modeling of Animatable Facial Expressions for High Density Models using
Geometry Images and Shader Programs
ın a
Benjam´ Hern´ndez, Isaac Rudomin
10:55-11:15 Multi-level Performance-driven Stylised Facial Animation
Fabian Di Fiore, Frank Van Reeth
11:15-11:35 Template-Model Based Modeling and Animation of Human Bodies with Anatomical Structure
Xiaomao Wu, Lizhuang Ma, Ke-Sen Huang, Yan Gao, Zhihua Chen
11:35-11:55 Estimation of Human Skeleton Proportion from 2D Uncalibrated Monocular Data
En Peng, Ling Li
Session F4: Animating Geometrical Models
Chair: Qunsheng Peng
14:10-14:35 Progressive Mesh Metamorphosis
Chao-Hung Lin, Tong-Yee Lee, Hung-Kuo Chu, Chih-Yuan Yao
14:35-15:00 Mesh Morphing Using PolyCube-Based Cross-Parameterization
Zhengwen Fan, Xiaogang Jin, Jieqing Feng, Hanqiu Sun
15:00-15:25 Physically Based Morphing of Point-sampled Surfaces
Yunfan Bao, Xiaohu Guo, Hong Qin
15:25-15:50 Mesh Decomposition Using Motion Information From Animation Sequences
Tong-Yee Lee, Ping-Hsien Lin, Shaur-Uei Yan, Chun-Hao Lin
Session F5a: Natural Phenomena and Special Effects
Chair: Nadia Magnenat-Thalmann
16:05-16:30 Extending the Photon Mapping Method for Realistic Rendering of Hot Gaseous Fluids
Byungkwon Kang, Insung Ihm, Chandrajit Bajaj
16:30-16:55 Controllable Local Monotonic Cubic Interpolation in Fluid Animations
Insung Ihm, Deukhyun Cha, Byungkwon Kang
16:55-17:20 Dynamic Modeling and Rendering of Grass Wagging in Wind
Changbo Wang, Zhangye Wang, Qi Zhou, Chengfang Song, Yu Guan, Qunsheng Peng
17:20-17:45 Blob-based Liquid Morphing
Xiaogang Jin, Shengjun Liu, Charlie C.L. Wang, Jieqing Feng, Hanqiu Sun
17:45-18:10 Animating Smoke with Dynamic Balance
Jin-Kyung Hong, Chang-Hun Kim
Session C4: Intelligent Agents
e
Chair: St´phane Donikian
14:10-14:30 Large-Scale Agent Formations in Virtual Environments Using Linear Elastic Shapes
Yi Li, Kamal Gupta
14:30-14:50 NeXus: Behavioural Realism in Mixed Reality Scenarios through Virtual Sensing
Gregory M. P. O’Hare, Abraham G. Campbell, John W. Stafford, Robert Aiken
14:50-15:10 Emotions, personality and social interactions modelling in a multiagent environment
e e
Alexis N´d´lec, Cyril Septseault, Dominique Follut, Gaele Rozec
15:10-15:30 Agent Paint: Intuitive Specification and Control of Multiagent Animations
Erik Millan, Isaac Rudomin
15:30-15:50 Ubiquitous Realities through Situated Social Agents
Mauro Dragone, Thomas Holz, Gregory M. P. O’Hare, Brian Duffy
16:05-16:25 Intentional Embodied Agents
Alan Martin, Gregory M. P. O’Hare, Bianca Schoen, John F. Bradley, Brian Duffy
16:25-16:45 Knowledge Exchange between Agents in Real-Time Environments
Jeppe Revall Frisvad, Peter Falster, Gert L. Møller, Niels Jørgen Christensen
16:45-17:05 A Model for Generating Competitive Groups Based on Virtual Agents Progression
Marta Villamil, Soraia Musse, Luis Luna de Oliveira
17:05-17:25 Using Social Agents to model a Virtual Character for the Recommendation of Students Tutors
Elisa Boff, Eliseo Berni Reategui, Rosa Maria Viccari
17:25-17:45 Using progressive adaptability against the complexity of modeling emotionally influenced vir-
tual agents
e
Ricardo Imbert, Ang´lica de Antonio
Day 3: Oct. 19
Invited Talk
09:00-10:00 Orientations and Rotations: A Coordinate-Free Pespective
Jehee Lee
Session F5b: Natural Phenomena and Special Effects
Chair: Weidong Geng
10:15-10:40 Interactive Venation-Based Leaf Shape Modeling
Sung Min Hong, Richard Bruce Simpson, Gladimir V. G. Baranoski
10:40-11:05 Real-time Simulation of Watery Paint
Tom Van Laerhoven, Frank Van Reeth
11:05-11:30 Real-Time Cartoon Animation of Smoke
Haitao He, Duanqing Xu
Session F6: Image, Color and Illumination in Animation
Chair: Pascal Volino
11:30-11:55 Image and Video Retexturing
Yanwen Guo, Jin Wang, Xiang Zeng, Zhongyi Xie, Hanqiu Sun, Qunsheng Peng
11:55-12:20 Enhanced Auto Coloring with Hierarchical Region Matching
Jie Qiu, Hock Soon Seah, Feng Tian, Quan Chen, Zhongke Wu
12:20-12:45 Deferred Shadowing for Real-Time Rendering of Dynamic Scenes Under Environment Illumi-
nation
Naoki Tamura, Henry Johan, Tomoyuki Nishita
Session C5: Animation Techniques
Chair: Taku Komura
10:15-10:35 Vectorization of Raster Line-Drawings in Cartoons
Quan Chen, Zhongke Wu, Feng Tian, Hock Soon Seah, Jie Qiu, Konstantin Melikhov
10:35-10:55 Geometric Detail Mapping for Point-sampled Geometry
Chunxia Xiao, Jieqing Feng, Tingfang Zhou, Wenting Zheng, Qunsheng Peng
10:55-11:15 A Fast Rendering Technique of Transparent Objects and Caustics
Kei Iwasaki, Fujiichi Yoshimoto, Yoshinori Dobashi, Tomoyuki Nishita
11:15-11:35 Watermarking Motion Clips
Shuntaro Yamazaki, Masaaki Mochimaru, Takeo Kanade
11:35-11:55 Exploring Offense-Defense Relationship for Chinese Martial Arts
Cong-Kai Lin, Jen-Yu Peng, Wen-Kai Tai
Event Venue
Registration and Coffee breaks 16/F Reception area, Li Ka Shing Tower (M Core)
Keynotes and Invited Talk Senate Room M1603, Li Ka Shing Tower
Full Paper Sessions: F1, F2, F3, F4a, F4b, F5, F6 Senate Room M1603, Li Ka Shing Tower
Conference and Poster Paper Sessions: C1, C2, C3, C4, C5, PT Room M1502, Li Ka Shing Tower
Lunch 14/F Chinese Restaurant, Li Ka Shing Tower
Reception Staff Restaurant, 4/F Communal Building
Banquet Wu Kong Shanghai Restaurant (Address: Basement, Al-
pha House, 27 Nathan Road, Tsim Sha Tsui, Kowloon.
Tel: 2366 7244)