Final program is available
http://www.alexandra.dk/dk/lige_nu/arra ... of-3d.aspx
10:00-10:20: Introduction
Jesper Mosegaard, Head of Computer Graphics Lab, the Alexandra Institute
10:20-11:00: Challenges and emerging methods in simulation and data analysis
Constructing computerized models of real world phenomena is always a challenge. Simply representing something continuous as a set of discrete elements (e.g. FEM) introduces error. Recent technological advances and ideas have addressed this issue, one being the moving mesh technology which limits the tradeoff between representation re-meshing and smoothing. This talk will address some of the very core issues that this technology can solve, and show and discuss some of the fields of application from fluid simulation to medical image analysis and classical neurology through neuroimaging.
Sune Darkner, Assistant professor, DIKU
11:00-11:30: Finally grape fruits and potatoes look nice: Fast and accurate rendering of translucent materials
Rendering translucent materials using Monte Carlo ray tracing is challenging. For such materials, rendering time often becomes prohibitively long due to the large number of subsurface scattering events. A faster approach is to use an approximate analytical model. However, the efficiency of the analytical models comes at the cost of losing translucency effects in the rendered result. In this talk, we present a new model for subsurface scattering, which is fully analytic, requires no precomputation, and is able to capture translucency effects accurately.
Jeppe R. Frisvad, Associate professor, DTU, Toshiya Hachisuka, Assistant professor, Aarhus University, and Thomas Kim Kjeldsen, Research and innovation specialist, the Alexandra Institute
11:30-12:30: Faithful digitization of large scenes, objects and people
While it is pretty simple to scan and digitize a paper document or even a complete book, there is no existing technology today that allows creating automatically a faithful digital representation of a given 3D object. It becomes even more complex for large environments or non-rigid shapes. Last but not least, why shouldn’t it be possible to digitize and structure even more abstract means, things that we cannot see or touch, such as user movements and even the skill of an expert? In this talk we will present an overview of the work carried out at the „Augmented Vision“ department at the University of Kaiserslautern and DFKI, Germany. The focus of our activities lies on digitization technologies, including scanning of objects, scenes and people. The technologies range from capturing of user activity and expertise based on visual-inertial on-body sensors over scene reconstruction based on high-resolution spherical HDR-images to the design of the first full spherical object scanner. The use of the captured information in the context of novel cognitive user interfaces based on augmented reality is also shown. Current results are presented and discussed in relation to the achieved quality and up-coming challenges.
Prof. Dr. Didier Stricker, Augmented Vision, DFKI
12:30-13.15: Lunch
13:15-14:00: Elementacular - a new generation of Maya plugin for interactive modeling and instant visualization of volumetric clouds and rocks
In an interdisciplinary cooperation between the Computer Graphics Lab at the Alexandra Institute, Sunday Studio and Javira we are developing a new kind of 3D content creation tool that enables creative professionals to work faster and more directly and intuitively with difficult and complex objects such as clouds and rocks. We achieve this through direct integration into existing pipelines, real time visual feedback and user-guided procedural generation.
Christian Esbo Agergaard, Technical director, Sunday Studio, and Jesper Børlum, Research and innovation specialist, the Alexandra Institute
14:00-14:30: How to build your own 3D scanner
3D scanners have a plethora of applications, from object digitization, digital modeling and robot navigation to gesture recognition and object tracking. With the advent of low cost depth-cameras like Microsoft Kinect and time-of-flight cameras, 3D scanners have become broadly accessible. These commercial offerings have specific trade-offs between scanning precision, speed, and field of view. By building your own device, you can control this trade-off and optimize for the problem at hand. In this talk, we will examine the techniques underlying optical 3D scanning with structured light, and present our performance optimized real time implementation. It is also shown, how this enables us to make better MRI medical images, which is the ultimate goal of our project.
Jakob Wilm, PhD student, DTU
14:30-15:00: Coffee break
15:00-15:30: Behind the animation short: “Rob ‘n’ Ron”
The talk will offer an insight into Tumblehead’s recipe for price winning animated short films: hack’n’slash,
considered pipeline solutions and pure coincidence. Bang! Bang!
Peter Smith and Lars Ellingbø, Tumblehead
15:30-16:00 State of the art in handheld 3D color scanning
In this presentation we show the scanning capabilities of the new TRIOS Color intraoral scanner. The scanner is used by dentists to capture the patients’ teeth geometries without the need for traditional impression taking. We will demonstrate how the scanner can capture natural looking colors that closely resemble the patients’ teeth. We then continue to show how the colors add significant value to the scanner, and how high quality textures and renderings can be used in dental CAD applications.
Peter Dahl, Software developer, 3Shape 