This is a special interest seminar series in computer graphics, of Kobe Studio Seminar for Studies. Each seminar has been planned by the organizers of KSS and has had discussions with the smallest possible number of participants.
For further information, please visit the following page.
We have organized several seminars for 2 years since Oct, 2013, and now have many participants who have projects related to Computer Graphics. We think that our first aim (discussions among researchers in different fields) is almost achieved hence each seminar of this special series is hosted in the main seminar series directly since January, 2016. Please visit the list of related seminars since January, 2016 to find the upcoming and past seminars of related topics.
This seminar is supported by Pixar. For more information, please see "Pixar's Contribution of RenderMan Licenses".
Moreover, please note that this seminar is supported by several external coordinators and supporters.
Please also visit the list of related seminars since January, 2016.
The local induction equation is well-known as a model of vortex filament in fluid dynamics. This describes a deformation of space curves in the binormal direction, and the complex curvature is governed by the nonlinear Schrödinger equation which is a typical integrable system. In this talk, starting from a brief introduction of discretization preserving integrability, we present a discrete model of dynamics of vortex filament based on this idea. The discrete complex curvature is governed by the discrete nonlinear Schrödinger equation (discrete in both space and time) proposed by Hirota-Tsujimoto. We discuss the relationship with the doubly discrete isotropic Heisenberg chain proposed by Hoffmann which is used for simulation of fluid flow in CG.
This talk is co-organized with KSS for Mathematics.
The project "Modeling for CG based on Discrete Integrable Systems" was started with this talk as a turning point, and had a workshop at IMI which achieved a very expansive discussion with specialists of visual effects in Japanese animation and film-making. Their presentation files are available at the hyperlink destination though in Japanese.
We would like to have a deep discussion on some topic related to the project: Finding Mathematical Structures in Computer Graphics, hence attendance is limited to only the related colleagues who have been contacted with us in advance. If you are a specialist in this topic and interested in our seminar, please contact to the corresponding organizer: Kosaku Nagasaka in advance.
We would like to have a study meeting on some topic in Computer Graphics, hence attendance is limited to only the related colleagues who have been contacted with us in advance. If you are a specialist in this topic and interested in our seminar, please contact to the corresponding organizer: Kosaku Nagasaka in advance.
From the moderators' point of view, we will give a short introduction to symbolic-numeric computation briefly.
We would like to have a deep discussion on some topic related to the project: Methodologies of Visual Appearance and Rendering, hence attendance is limited to only the related colleagues who have been contacted with us in advance. If you are a specialist in this topic and interested in our seminar, please contact to the corresponding organizer: Kosaku Nagasaka in advance.
We would like to have a deep discussion on some topic related to the project: Finding Mathematical Structures in Computer Graphics, hence attendance is limited to only the related colleagues who have been contacted with us in advance. If you are a specialist in this topic and interested in our seminar, please contact to the corresponding organizer: Kosaku Nagasaka in advance.
A key aspect of Non-Photorealistic Rendering (NPR) is artistic stylization as a practical creative tool. We first introduce our past challenges to relax physical constraints of photorealistic rendering for a case of simple quantized cartoon shading styles. We explored both 2D and 3D approaches to integrate artistic stylization into the formulation of photorealistic rendering. In the latter of talk, we would like to discuss how to establish further stylized appearance and animation related to the project: Methodologies of Non-Photorealistic Rendering. We survey several stroke rendering techniques to provide inspiration for a future research direction.
From the moderators' point of view, we have done briefly a short survey on architectural design and film making. In this talk, we will give a short introduction on this survey briefly.
This paper presents a method that employs parametric surfaces as surface geometry representations at any stage of a computational process to compute self-supporting surfaces. This approach can be differentiated from existing relevant methods because such methods represent surfaces by a triangulated mesh surface or a network consisting of lines. The proposed method is based on the theory of Airy stress functions. Although some existing methods are also based on this theory, they apply its discrete version to discrete geometries or directly determine stress functions through elementary mathematical functions. Conversely, the proposed method applies the theory to parametric surfaces directly and the discrete theory to the edges of parametric patches simultaneously. The discontinuous boundary between continuous patches naturally corresponds to ribs seen in traditional vault masonry buildings. We use NURBS surfaces in this study; however, the basic idea can be applied to other parametric surfaces. A variety of self-supporting surfaces obtained by the proposed computational scheme is presented.
Starting from form-finding problems of tensegrity structures, I'll give a quick introduction of FEM, differential geometry and principle of virtual work.
We would like to have a deep discussion on some topic related to the project: Finding Mathematical Structures in Computer Graphics, hence attendance is limited to only the related colleagues who have been contacted with us in advance. If you are a specialist in this topic and interested in our seminar, please contact to the corresponding organizer: Kosaku Nagasaka in advance.
We would like to have a deep discussion on some topic related to the project: Finding Mathematical Structures in Computer Graphics, hence attendance is limited to only the related colleagues who have been contacted with us in advance. If you are a specialist in this topic and interested in our seminar, please contact to the corresponding organizer: Kosaku Nagasaka in advance.
We would like to have a deep discussion on some topic related to the project: Finding Mathematical Structures in Computer Graphics, hence attendance is limited to only the related colleagues who have been contacted with us in advance. If you are a specialist in this topic and interested in our seminar, please contact to the corresponding organizer: Kosaku Nagasaka in advance.
We would like to have a deep discussion on some topic related to the project: Finding Mathematical Structures in Computer Graphics, hence attendance is limited to only the related colleagues who have been contacted with us in advance. If you are a specialist in this topic and interested in our seminar, please contact to the corresponding organizer: Kosaku Nagasaka in advance.
It is often useful to compute a lot of matrix exponentials in computer graphics (CG). The exponentials of a matrix is used for the smooth deformation of 2D or 3D meshed CG objects. We introduce a simple formula for the matrix exponential of a 3x3 real symmetric matrix using Viete's Formula. We incorporated our implementation into a shape deforming tool developed by Kaji et. al. This talk is based on joint work with K. Matsushita.
This is a closing report and post discussion on KSS Projects for Communication: Low Dimensional Topology and Computer Graphics, and we would like to have a deep discussion on the topic hence attendance is limited to only the related colleagues who have been contacted with us in advance. If you are a specialist in this topic and interested in our workshop, please contact to the corresponding organizer: Kosaku Nagasaka in advance.
Evolution Equations (EE for short) have been vigorously studied in views of mathematics as well as of physics ever since the notion of EE was introduced to describe the evolution of physical systems in an abstract fashion. In particular, the theory of EE has been applied to make a great contribution to explore the study of Partial Differential Equations. Moreover, in this field, Japanese mathematicians such as Kosaku Yosida, Toshio Kato, Yukio Komura established many pioneer works.
In this talk, the basic idea of EE will be explained, and then, the development of this field so far will be briefly surveyed (within the limits of time and speaker's ability). In the latter half, we shall review speaker's recent works, particularly on Variational Principle for EE with energy dissipation structures, to discuss some philosophical aspect, basic idea and results obtained so far. The latter half is based on a series of joint works with Ulisse Stefanelli (Vienna).
This talk is co-organized with KSS for Mathematics.
In this talk, we present our recent work on real-time rendering under all-frequency environment light with highly glossy materials, and our recent work on visual simulation of melting ice phenomena.
Recent advances on image-based lighting methods have revealed that spherical Gaussians have several nice properties for rendering, and are suited to represent all-frequency signals compactly and accurately such as complex environment lighting and highly glossy BRDFs. In this talk, we present two techniques based on spherical Gaussians, a real-time image-based lighting method for dynamic scenes and a bi-scale material editing method.
In addition, we introduce a particle-based method for the simulation of the melting and freezing of ice objects and the interactions between ice and fluids. We demonstrate our simulation and rendering method for depicting melting ice at interactive frame-rates.
On November 22, 2014, we introduced some methods for the orientation problem. In this talk, we discuss the study of it.
Attendance is limited to only the related colleagues who have been contacted with us in advance. Please contact to the corresponding organizer: Kosaku Nagasaka in advance.
In this talk, we will focus on the talk at the previous seminar, and discuss about some possibilities on the 2D orientation problem from the viewpoint of our project: Symbolic-Numeric Computations for Computer Graphics, KSS Projects for Communication.
Attendance is limited to only the related colleagues who have been contacted with us in advance. Please contact to the corresponding organizer: Kosaku Nagasaka in advance.
In this talk we exhibit some examples of how computer graphics have contributed to the development of minimal surface theory.
This talk is co-organized with KSS for Mathematics.
We'd like to have discussions on Prof. Fujimori's past studies in minimal surfaces and CG. Please visit the following link (which may be helpful) if you'd like to survey the topic after the seminar.
This discussion is co-organized with KSS for Mathematics.
We will have a small discussion followed by Workshop: Around Mathematical Physics and Geometry with its participants. If you are interested in this discussion, please contact to the corresponding organizer: Kosaku Nagasaka in advance.
We'd like to discuss the following topics:
Please visit the following link (which may be helpful) if you'd like to survey the topic after the seminar.
Light transport simulation in computer graphics is imperative for photorealistic image synthesis. Light transport simulation based on geometric optics can be formalized as a solution to the integral equation. This integral equation, however, has no analytical solution in general cases. We thus need to use a numerical method such as Monte Carlo methods to obtain its approximate solution. In this talk, I will cover the contents of the tutorial for photon density estimation methods which I organized for past SIGGRAPH and SIGGRAPH Asia conferences. In particular, I will explain the progressive photon density estimation framework which is known as the first practical framework which is capable of simulating luminaires from the filaments/LEDs level. I will also explain some of the recent work in light transport simulation, in order to communicate mathematical ideas behind those work.
In this talk, we discuss the effectiveness and disadvantages of our algorithm which is completely different from but based on the algorithm introduced in the past talk.
Moreover, there are two related talks: Kobe Studio Seminar for Mathematics.
In this talk, we consider the following two topics related to computer graphics and mathematics: "Filament-based smoke and integrable system" and "Conformal deformation of triangle meshes and Dirac operator". We explain mathematics backgrounds (first part) and implementations (second part).
Filament-based smoke and integrable system:
The relationship between the motion of vertex filaments and the nonlinear Schrodinger equation is discovered by Hasimoto (1972). The nonlinear Schrodinger equation is an important example in the so-called "integrable system". We introduce the Steffen Weißmann's Ph.D. thesis "Filament-Based Smoke", and discuss applications of integrable systems to computer graphics.
Conformal deformation of triangle meshes and Dirac operator:
The conformal structure is important in mathematics and is described by the Dirac operator. Also, the conformal structure in dimension two or three is important in computer graphics. For example, deformation of objects and texture mapping. We introduce the Keenan Crane's Ph.D. thesis "Conformal Geometry Processing", and consider the relationship between the conformal deformation of triangle meshes and the Dirac operator. In particular, we discuss a discretization of the Dirac operator.
This talk is co-organized with KSS for Mathematics. We will also have some related discussions from the mathematical and algorithmic points of view.
Breadth-first search is the one of most important and fundamental graph processing and has many applications, such as connected component, centrality metrics, and maximum flow. Fast computation of BFS is not easy to that it has a low arithmetic intensity and many irregular memory accesses with poor reusability. Thus, we proposed the efficient algorithm which considers the memory hierarchy in a computer. Our algorithm achieved the fastest entry on single-node computer and the first position of energy-efficient in current Graph500 and Green Graph500 list.
This talk is co-organized with KSS for Mathematics.
Some algorithm requires frequent accesses for widely memory spaces, such as large-scale graph computation in the computer graphics field. In such an algorithm, theoretical complexity and performance on real computer do not necessarily go together. In this talk, we mention the fast computation which considers the memory hierarchy in a computer, and explain the gap of theory and practice.
This talk is co-organized with KSS for Mathematics.
We consider a simple class of Kleinian groups called once punctured torus groups. In this talk, we will show how to create a computer program from scratch that can visualize fundamental domain and limit sets of the groups. As an application, we will show the computer generated pictures of discreteness loci of several slices of SL(2,C)-character variety.
We consider a simple class of Kleinian groups called once punctured torus groups. In this talk, we will show how to create a computer program from scratch that can visualize fundamental domain and limit sets of the groups. As an application, we will show the computer generated pictures of discreteness loci of several slices of SL(2,C)-character variety.
Please make sure that the participants may have read the references at the "Color imaging" on https://sites.google.com/site/yusukekiriu/home/renderman2012. Moreover, we may have a discussion on fluid simulation and compositing hence we encourage you to read the following websites at Sidefx: http://www.sidefx.com/index.php?option=com_content&task=view&id=2200&Itemid=344 and Pixar: http://renderman.pixar.com/resources/current/rms/ice_Scripting_Environment.html.
Other references: Paperman; http://www.disneyanimation.com/projects/paperman. Physically Based Lighting at Pixar/Christophe Hery, Ryusuke Villemin. Stylizing Animation By Example/Pierre Benard et al. Fluid Simulation for Computer Graphics / Robert Bridson. Animating Sand as a Fluid/ Y. Zhu, R. Bridson.
We introduce computational invariant theory and ask whether or not there is any relationship between computational invariant theory and computer graphics. We welcome everyone who can discuss the relationship.
Creating realistic images has been a major focus in computer graphics. From my point of view, an issue is a "relationship" between invariant theory and the Rendering equation(J.Kajiya, 1986).
We discuss inverse rendering problems, voxel cone tracing, multiple importance sampling, optimization theory, etc. to study the trade-off between the rich appearance and efficient calculation for the design of a shading pipeline.
We also explain the details of Renderman Shading Language and several approaches of shading models used in our discussion.
We introduce a method to construct, in 3-space, surfaces whose boundary is knotted in the specified way. We also show a continuous sequence of surfaces spanning fixed circles. These would be interesting examples of Mathematical objects to visualize by RenderMan.
First, we will talk about a brief explanation of Renderman Shading Language, Renderman Interface Bytestream Protocol as a course of Entry level.
Second, several styles of production pipelines to combine typical use cases of production pipeline in Japan.
Finally, we will explain our current study "physically plausible shading for non-photorealistic shading with several motion blur techniques" with the brief explanation of several tech-memos and papers of algorithms of texture synthesis method of Image Analogies as an example of production pipeline.
references: https://sites.google.com/site/yusukekiriu/home/renderman2012, universal scene description, gaffer