Spring 2006

VISUALIZATION FRIDAY FORUM

Fridays 12-1pm LSRC D106

Lunch Served


Jan 27 - Visual Effects and Animation at Sony Image Works

Jeff Dillinger
Sony Image Works

Jeff Dillinger will present an overview of life working on both a live action visual effects project as well as an animated feature. He will cover the responsibilities of individuals in each department and describe some of the issues that they face. Examples will include the challenge of creating imagery that is physically correct versus what looks correct to the viewer.


Feb 3 - High Resolution X-Ray: Digital Subtraction Angiography for Small Animal Imaging

Ming De Lin
Dr. Allan Johnson's Radiology Group

The availability of genetically altered rodent models of human disease and the increasing use of small animals in basic research have spurred extraordinary interest in new imaging methodologies—particularly magnetic resonance microscopy, microCT, and microPET.

There exists enormous potential for functional imaging using micro X-ray given the ease of use and its potential speed. Functional imaging in rodent models can be addressed particularly well through the use of digital subtraction angiography (DSA). We describe here a system designed explicitly for DSA in the rodent and the optimization of acquisition parameters required to perform the highest quality functional subtraction angiograms.


Feb 10 - Croquet - A Platform for Collaboration

Dr. Julian Lombardi and Dr. Marilyn Lombardi
OIT

Croquet utilizes 3D to act as the social dimension of interaction. The system is focused on interactions inside of a 3D shared space that is used for context based collaboration, where each user can see all of the others and what their current focus is. This allows for an extremely compelling shared experience. A new collaboration architecture/protocol called TeaTime has been developed to enable this functionality. The rendering architecture is built on top of OpenGL.

Croquet makes it possible to change and author virtual worlds in collaboration with others inside them while they are operating. Croquet users, with appropriate sharing privileges, may visit and work inside any other Croquet-based world on the network. The system provides a software environment where anything can be created or represented; where everything can be modified; where development and user environments are one in the same.

Julian will present the current state of the system, provide a number of working example applications and environments, and discuss how TeaTime and the requirements of a 3D collaborative object model changes the dynamics of development. The future direction of the ongoing open-source effort will be laid out and discussed.


Feb 17 - Dynamic Data interfaces with Flash and XML (Presentation - (Quicktime); Flash Demo )

Casey Alt
ISIS

Casey Alt, the Administrative Director of Information Science and Information Studies (ISIS), will present an introductory overview to building interactive, rich-media (dynamic vector graphics, bitmaps, audio, and video) interfaces with Macromedia/Adobe Flash and XML (Extensible Markup Language). The combination of Flash and XML allows designers to rapidly develop and deploy dynamic online data interfaces with both asynchronous and synchronous client-server data interactions. Casey will briefly discuss the basic concepts and capabilities of Flash and XML then will walk a few different sample visualizations. Casey has been developing applications with Flash and XML since 2000.


Feb 24 - Three dimensional hepatobiliary architecture in medaka (Oryzias latipes)

Ron Hardman

Nicholas School Of the Environment

Using non invasive in vivo imaging and a living small fish animal model we have elucidated the 3 dimensional architecture of the lower vertebrate tubular liver phenotype. These investigations suggest that tubular hepatic architecture in medaka (Oryzias latipes) is based on a polyhedral, primarily hexagonal, motif, and that the biliary system is a parallel network, as opposed to serial or linear based biliary tree model.


Mar 3 - Creating a Virtual Storyspace

Michael Nitsche
Georgia Tech

The Mindstage project grew out of earlier work on the integration of narrative and mediated virtual architecture. The basic research question was whether a Real-Time 3 Dimensional Virtual Environment's spatial structure can utilized for educational purposes. What would be a sufficient architectural design of virtual space that could reinforce the piece’s impact and promote engagement, exploration and memorability?

A virtual three-dimensional space was created as a performance space for a given educational content: a traditional linear lecture. The lecture chosen for the prototype is an introduction into film design delivered by the award- winning film designer Christopher Hobbs. The topic of the talk defined the environment and its functionality. This talk will look into the design issues and the practical implementation of the prototype of Mindstage as it was delivered in 2004.


Mar 10 - Visualizing Dynamic Data in Realistic 3D Tissue Architecture Using AVS Express

Sarah Roberts
G-BME

An increasing number of investigators are using computational modeling of physiological phenomenon as an important supplemental research tool. Thus, it becomes necessary to visualize the in silico data in a meaningful context that can be easily analyzed by the researcher, as well as, presented to others. Advanced Visual Systems (AVS/Express) software provides a customizable format for visualizing an array of data formats from simple 2D graphs to complex 3D multivariate information. One specific area of research that can exploit the advantages of the AVS software is in the study of cardiac electrophysiology. Given a uniquely defined finite volume mesh describing cardiac tissue spaces in silico, AVS can map respective time-variant voltage data to specific tissue architecture.


Mar 17 - Spring Break
NO FRIDAY FORUM


Mar 24 - Image Processing Tools for Improved Data Visualization

Dr. Rebecca Willett
ECE

In this talk, Dr. Willet will present an overview of a several image processing and feature extraction methods which can be used to prepare data for effective visualization. In particular, she will describe several denoising and restoration techniques which can be used to remove noise and other sensing artifacts from image data, and highlight the the regimes in which each method is most effective. She will also review several feature extraction methods which can be used in conjunction with many visualization techniques.


Mar 31 - Putting Things In A Good Light

Dr. Michael Gustafson

ECE

"Barbara's little looking-glass hung in a good light near the window..." - Charles Dickens, "The Old Curiosity Shop"
Since their invention in the 1960's, the humble light-emitting-diode (LED) has been used for ever-more interesting applications and available in an ever-widening array of colors. In this talk, Dr. Gustafson will demonstrate the use of an array of pulse-width dimmable LEDs to produce a near-infinite combination of light colors. The array will also be used in conjunction with image processing techniques in Matlab to produce appropriate lighting for still and animated images - thus putting them in a good light.


Apr 7 - Tools to Navigate in 7 Dimensions

David Richardson

Biochemistry

Two of the research problems in our lab involve data best described by 7 parameters: the local backbone conformations in RNA structures, and the relative positioning of loop endpoints in proteins. To ease our struggle with exploring and analyzing these data sets, we have recently developed two new tools that let us interactively visualize and annotate the clustering of data points in all 7 dimensions. The talk will demonstrate what this process is like.


Apr 14 - Data-driven models of 3D crown structure and light in closed forest canopies

Michael Wolosin

Biological Sciences

This two-part talk will focus first on the methods I've used to model three dimensional tree crown structure based on field and remote sensing data, and second on a few visualization techniques I'm using to display the results of my research. Overlapping frames from plane-based digital videography of Duke Forest field sites were used both to extract a set of 3D points representing the positions of foliage in the forest canopy, and to visually segment the forest canopy into individual tree crowns. I then combined the interpreted canopy points, crown polygons, and field measurements of tree locations and heights into models of the 3D structure of over 900 trees in our field plots. These virtual trees were rendered in a hardware analog of a ray-tracing light model to estimate the integrated annual light levels throughout the canopy volume. This project lends itself naturally to several visualization techniques, which I use to convey both the methods of analysis and the results of the study. These methods include animations and virtual "fly-throughs," slices extracted from data cubes, and the use of color and shading to convey different types of information visually and to tie together different parts of the study.


Apr 21 - Morphometric Analysis of Primate Sensory Adaptations Using High Resolution Computed Tomography

Michael Malinzak
BAA

High-resolution computed tomography (CT) permits extensive and non-destructive study of both recent and fossilized primate cranial specimens. Special sensory adaptations are among the most salient and conserved anatomic features of the primate cranium. Using CT-based comparative methods, we seek to better understand the functional and evolutionary basis of morphologic variation with particular regard to the visual and vestibular systems. In this two-part talk, I will present recent collaborative findings on the visual system of an early South-American primate as well as my ongoing work on the functional significance of primate labyrinthine anatomic diversity.

For more information, please contact
Rachael Brady

Organized by
Computational Science, Engineering, and Medicine (CSEM),
Visualization Technology Group (VTG), and
Information Science + Information Studies (ISIS)