The Visualization Friday Forum

Fall 2005 * Noon-1pm * D106 LSRC

Sept 9 - Visualization at Duke
Rachael Brady
Visualization Technology Group

Visualization applies the algorithms of computer graphics with the fields of perception and representation to communicate digital information visually. Visualization is used for presentations, art creation, data analysis, model validation, illustration, data exploration, entertainment, and cognitive studies.

I'll start the fall term by introducing the Visualization Technology Group and the Friday Forum series.  In particular, I'll fill you in on some of the happenings from over the summer... such as learning a VR Authoring system for the DiVE (nee Visroom) and talk about some of the projects planned for that space.  CSEM's new quarters on 3rd floor of North building has opened, and the cluster has been moved to a room with cooling and power.

Sept 16 - Image and Meaning (2) [PowerPoint Presentation]
Rhazes Spell
Visualization Technology Group

The purpose of the Image and Meaning initiative is to: '...help scientists, writers and visual communicators  develop and share improved methods of communicating  scientific concepts and technical information through  images and visual representations linked to appropriate  text. The goal is to enhance the level of discourse  within the scientific community, among teachers and  those who communicate with the public through the mass media. '

This summer I had the opportunity to participate in IM2 (Image and Meaning 2), the second installment of Image and Meaning.  For this Friday forum, I will recap the events from the conference that are of interest to the Duke Visualization community.

Sept 23 - A New Approach to Introductory Programming -
Building Virtual Worlds with Alice [PowerPoint Presentation]
Susan Rodger
Computer Science

The software tool Alice makes learning to program easy and fun.  With Alice a novice can create a 3D virtual world.  This world might be a story, or it might be an interactive game.  Students select 3D objects, place them in a virtual world, and give these objects functionality. For example, the world may be a snow scene with a frozen pond, an object may be an ice skater, and the student teaches the ice skater how to skate, spin and even blink. Alice cuts down on frustration with its drag-and-drop interface. It is difficult to make a typing or syntax error.  We present the tool Alice and show how we use it for teaching introductory programming in the course CompSci 4 at Duke.

Sept 30 - Quick and Dirty Visualization for Geometric Algorithms [PDF Presentation]
Jeff Phillips
Computer Science

I have been working on several computational geometry algorithms for matching various datasets. The purpose of my research is not to visualize the algorithm, but rather to understand the properties of the algorithms and find improvements. Yet, I felt I still needed a visualizer for several reason: 1) raw numbers from these geometry algorithms are hard to interpret, 2) I want to be able to verify whether the code is correctly implemented, 3) when I try various potential improvements, I want to know immediately whether they work without tedious data analysis. The result is a quick and dirty visualization tool, which was simple to implement, yet served my purposes. As an added benefit, with this tool, it is much easier to explain my work and its contribution.

Oct 7- A Trip in the Virtual Forest –
Building Realistic Forest Domains for Atmospheric Large Eddy Simulations [PowerPoint Presentation]
Gil Borher
Civil Engineering

Turbulence affects the onset, propagation and termination of a diverse array of biological transport phenomena, such as pathogen spread, biogenic aerosol spread, seed dispersal, and cloud interaction via biogenic cloud condensation nuclei. It also controls the exchange of water vapor and CO2 from plant canopies into the atmosphere. Turbulent eddies occur over a wide range of spatial and temporal scales and are strongly affected by structural changes in the landscape. Turbulence is chaotic with strong non-linear interdependence among multiple influential factors. Wind dispersal of seeds, and in particular long distance dispersal of large seeds, which is a rare event that happens in the "tail" of the dispersal distribution, can be highly biased by averaging the wind flow-field by using a coarse resolution model. Thus, dispersal processes might not be correctly modeled at a mesoscale resolution. Instead, a spatially explicit high-resolution large eddy simulation (LES) of heterogeneous canopies is needed to study these effects.

These types of LES virtual experiments use computer based high-resolution calculations of air-flow as surrogates for actual wind, which is too hard to experimentally manipulate or measure in detail. But what kind of detailed "virtual canopies" can be used by LES models as simulation domains? Is it possible to measure all the canopy features needed to reproduce a canopy in the virtual domain? And how can all the resulting highly detailed data be viewed?

The virtual canopy generator creates random canopies, with statistical features that are based on features of actual canopies. A step-by-step "virtual canopy" recipe and some preliminary application will be presented.

Oct 14 - The Use of Physiological Reaction in Virtual Environments
Sharif Razzaque
Computer Science, UNC

A user’s physiological signals, such as her heart rate and skin conductance, can provide insight into how she is being affected by a virtual environment(VE). Common measures of VE effectiveness include the extent to which they induce a sense of  presence (the feeling of being “in” the virtual scene) or co-presence (the feeling of being with others).One hypothesis is that the more a VE seems real, the more it evokes physiological responses similar to those evoked by the corresponding real environment.

I’ll present several experiments that have measured physiological signals during VE sessions, and used them as indicators of presence and co-presence.  I’ll discuss the results of those experiments, as well as strategies and pitfalls related to the use of physiological monitoring during VE experiences.

Oct 21 Redirected Walking
Sharif Razzaque
Computer Science, UNC

Locomotion in human-scale, immersive virtual environments can be specified by flying with a hand-controller, using a treadmill, walking-in-place, etc. Real walking, where the user actually and physically walks in the lab, and moves the same distance in the virtual scene, is better than flying. It is more input-natural, does not require learning a new interface, results in a greater sense of presence, and theoretically results is less simulator sickness.

With real walking, however, the size of the virtual scene is limited by the size of tracked area. For example, for an architect to really walk in a virtual prototype of a house, the tracked area must be as large as the house. This requirement makes real walking infeasible for many facilities and virtual scenes.

To address this limitation, I have developed Redirected Walking, which makes the user turn herself by interactively and imperceptibly rotating the virtual scene about her. Under the right conditions, the user would unknowingly walk in circles in the lab, thinking she is walking in a straight and arbitrarily long path in the virtual scene.

I'll describe Redirection, discuss its theoretical underpinnings, and argue that it can be used: 1) to make the user turn themselves 2) without causing the user to be aware of Redirection or modify her conscious behavior 3) without unacceptably increasing the level of simulator sickness, most importantly, 4) to useful effect:

A) In head-mounted display systems, the user can experience a virtual scene larger then the lab while also having the benefits of real walking.

B) In an open-backed, three-walled CAVE, users can have the increased presence and input-naturalness normally of a fully enclosed CAVE

I'll finally discuss continuing work on on Redirection to address the lack of haptic feedback (being able to touch and feel virtual objects) in large virtual scenes. 

Oct 28 - What can patent landscapes tell us about innovation?
Eric Giannella
ISIS

Patents provide an excellent opportunity for studying technological change, but the sheer amount of patent information can easily become overwhelming. How does one classify and analyze such a dataset without imposing biases? If most patents are junk, does that necessarily mean they are noise in our study? What types of conclusions can we draw about the history of a field from a landscape of patent information? I will demonstrate techniques for mapping and exploring patent bibliographic and citation data. I will also discuss pitfalls in analyzing citation maps and describe methods of layering visual data.

Nov 4 - Surface Parameterization: Flattening the protein-protein interface surface.
Madhuwanti Vaidya
Computer Science

I will give a brief overview of various surface parameterization techniques and then focus on flattening the protein-protein interface surface. This surface is a 2-manifold and we need a function that with minimum distortion maps the surface onto a convex polygon in the plane. We have looked at three convex combination mappings and have compared the three using area and angle distortion.

Nov 11 - Navigating through a Virtual World
William Sherman
Desert Research Institute, Reno, NV

Often virtual reality experiences are designed to mimic the real world, but sometimes this is not possible or even desireable. So when not constrained by a lifelike interface to the world the virtual world designer must consider how to enable the user to move from place to place. Equally important, the design must help the user move without getting lost. In this talk, we'll look at a variety of styles of travel through virtual reality worlds, along with methods of keeping the user oriented and heading in the right direction.

Nov 18 - Highlights from the IEEE Visualization conference, and the VR Jam Session
Rachael Brady
Visualization Technology Group

I will present some of my favorite papers from the IEEE Visualization conference.  Just to give you a taste -- one is a very clever way to lay out directed acyclic graphs  (i.e. complicated node-link diagrams).  Another is a semi-automated process to create textures for showing the shape of imbedded surfaces.  These are both very difficult  problems.  There are several more, both from Information visualization and on the regular visualization side.

I'll also talk about some of the applications that David Zielinski was able to get installed from UIUC and Brown.

Nov 25 - Thanksgiving
No Forum

Dec 2 - Calm Technology: Designing information with manners
(http://www.ambientdevices.com/cat/platform.html)
David Rose
Ambient Devices

THIS TALK WILL TAKE PLACE IN ROOM 240 OF THE FRANKLIN CENTER.

Wireless devices today require an enormous amount of a very scare resource—our attention. Their interfaces are inscrutable and we continuously apologize for their lack of social etiquette with interruptive ringing, beeping, and buzzing.

Dec 9 -