Bruno Levy is a Researcher with INRIA. He is the head of the ALICE
research group. He develops the "Numerical Geometry" approach,
i.e. new formalisms to define geometrical operators acting on
discretized objects. His main contributions concern texture
mapping and parameterization methods for triangulated surfaces.
He obtained his Ph.D in 1999, from the INPL (Institut National
Polytechnique de Lorraine). His work, entitled "Computational
Topology: Combinatorics and Embedding", was awarded the SPECIF
price in 2000 (best French Ph.D. thesis in Computer Sciences).
After his Ph.D., he did a Post-Doc in Stanford, in the SCCM Dept.
headed by G. Golub (applied math. and numerical analysis),
and in the Earh Sciences Dept. (with K. Aziz and A. Journel).
His main results have been transferred to the industry:
His parameterization method is the kernel of the gridding tools
available in the Gocad modeler, commercialized by the Earth
Decision Sciences company, and used by major oil companies.
His texture mapping algorithms are implemented in several
3D modelers, including Maya, Silo and Blender.
For more details:
Invited Talk: Laplace-Beltrami Eigenfunctions:
Towards an algorithm that "understands" geometry
One of the challenges in geometry processing is to automatically
reconstruct a higher-level representation from raw geometric data.
For instance, computing a parameterization of an object helps
attaching information to it and converting between various
representations. More generally, this family of problems may be
thought of in terms of constructing structured function bases
attached to surfaces.
In this paper, we study a specific type of hierarchical function
bases, defined by the eigenfunctions of the Laplace-Beltrami
operator. When applied to a sphere, this function basis corresponds
to the classical spherical harmonics. On more general objects,
this defines a function basis well adapted to the geometry and the
topology of the object.
Based on physical analogies (vibration modes), we first give an
intuitive view before explaining the underlying theory. We then
explain in practice how to compute an approximation of the
eigenfunctions of a differential operator, and show possible
applications in geometry processing.
Fujio Yamaguchi is a Professor in the Department of Mechanical
Engineering at Waseda University. He received his Doctorate in
Engineering in 1978. He was an Associate Professor at the Kyushu
Institute of Design from 1977 to 1986. As a visiting Associate
Professor, he taught computer graphics and computer-aided geometric
design in the Department of Computer Science, University of Utah,
in 1978 and 1979. His research interests include computer-aided
geometric design and computational geometry. He is an author of
many books in his area including "Curves and Surfaces in Computer
Aided Geometric Design" (1988, Springer) and "Computer-Aided Geometric
Design--A Four-Dimensional Approach" (2002,Springer).
For more details:
Invited Talk: Shape Modeling in a Homogenized Geometric Paradigm
The author proposes replacing conventional Euclidean processing
with homogeneous processing, which is free from the ill effects of
division and has the desirable properties of duality and projective
invariance. The division operation negatively affects such
properties as exactness, robustness, and simpleness in
computer-aided design systems. He mainly discusses two types of
geometric processing in the homogenized geometric paradigm, i.e.
subdivision method in surface design and set operations in solid
Alla Sheffer is an assistant professor in the Computer Science
department at the University of British Columbia (Canada). She
conducts research in the areas of computer graphics and computer
aided engineering. Dr. Sheffer is predominantly interested in the
algorithmic aspects of digital geometry processing, focusing on
several fundamental problems of mesh manipulation and editing. Her
recent research addresses algorithms for mesh parameterization,
processing of developable surfaces, mesh editing and segmentation.
In her earlier work she investigated generation and editing of
finite element meshes for engineering applications.
Alla Sheffer received her PhD from the Hebrew University of
Jerusalem in 1999. From 1999 till 2001 she was a postdoctoral
associate in the University of Illinois at Urbana-Champaign.
Between 2001 and 2003 Dr. Sheffer was an assistant professor at
Dr Sheffer is a program co-chair of the Symposium on Geometry
Processing 2006. Her current and past committee duties included
chairing the 10th International Meshing Roundtable and serving
on the program committees of SIGGRAPH, Eurographics, Shape Modeling
International, Solid Modeling, Symposium on Geometry Processing,
Graphics Interface, International Meshing Roundtable, Computer
Graphics International, and WSCG.
For more details see
Invited Talk: Shape Modeling for Dummies
The creation of novel 3D content is one of the major bottlenecks of
modern computer graphics. Commercial modeling systems are targeted
toward expert users and require significant time, expertise and artistic
talent to generate 3D shapes. Hence, much of the emphasis in recent research
is on simplifying model creation. One approach is to provide simpler interfaces,
such as sketching tools. Another recent trend focuses on reuse of existing
models by providing robust editing tools. Most of those tools still require
the user to have modeling expertise and non-negligible artistic ability.
In this talk I will review some recent trends in providing simple modeling
interfaces. I will then demonstrate that by narrowing the scope of the problem, and
focusing on modeling within a specific set of models, it is possible to develop
a true "modeling for dummies" interface which does not require any modeling
expertise or talent. Despite its simplicity the interface allows for creation
of rich geometric content within a matter of minutes. The proposed modeling
system, Shuffler, operates on sets of models that have a similar part-based
structure. Example sets include quadrupeds, humans, chairs, and airplanes.
Shuffler automates the process of creating new models by composing
interchangeable parts from different existing models within each set. As noted,
it does not require the users to perform any geometric operations; they simply
select which parts should come from which input model, and the system composes
the parts together. To enable this modeling paradigm, Shuffler precomputes
the part structure within each model and establishes the correspondences
between the interchangeable parts on different models.
The talk is based on joint work with Vladislav Kraevoy and Dan Julius.
Last Update: 7 April, 2006