| Visualization Brownbag Department of Architecture and Department of Computer Science Friday 12:30 - 1:30 P.M. Curved Room 414, Langford Center (Building C) |
|
|
|
|
|
|
June 13 - Streaming Surface Reconstruction Using Wavelets Abstract - We present a streaming method for reconstructing surfaces from large data sets generated by a laser range scanner using wavelets. Wavelets provide a localized, multiresolution representation of functions and this makes them ideal candidates for streaming surface reconstruction algorithms. We show how wavelets can be used to reconstruct the indicator function of a shape from a cloud of points with associated normals. Our method proceeds in several steps. We first compute a low-resolution approximation of the indicator function using an octree followed by a second pass that incrementally adds fine resolution details. The indicator function is then smoothed using a modified octree convolution step and contoured to produce the final surface. Due to the local, multiresolution nature of wavelets, our approach results in an algorithm over 10 times faster than previous methods and can process extremely large data sets in the order of several hundred million points in only an hour. Top May 30 - Gradient Space Projection Abstract - Gradient space projection is a simple method for handling projections on the hemisphere. It defines a coordinate system on the hemisphere surface that effectively converts the hemisphere to an infinite plane tangent to the top of the hemisphere, which is called the gradient space. Using this coordinate system, projections onto the hemisphere are replaced by simple perspective projections onto the gradient space. This approach totally eliminates the non-linearities caused by the spherical surface and permits exact hemisphere projection computations using only a single projection operation. I will also present different sampling techniques on the gradient space and provide qualitative comparisons. Top May 2 - Mesh Colors Abstract - Mesh colors is a revolutionary way of keeping color data on polygonal meshes. It is designed to replace texture mapping completely in applications that require one-to-one mappings (like 3D painting and baking ambient occlusion) for complicated models that are difficult to map and use with 2D textures. Mesh colors associate color data directly with the polygonal meshes. The approach eliminates all problems deriving from using a map from texture space to model space. Mesh colors is an extension of vertex colors where, in addition to keeping color values on each vertex, color values are also kept on edges and faces. Like texture mapping, the approach allows higher texture resolution than model resolution, but at the same time it guarantees one-to-one correspondence between the model surface and the color data, and eliminates discontinuities. We show that mesh colors integrate well with the current graphics pipeline and can be used to generate very high quality textures. http://www.cemyuksel.com/research/meshcolors/ Top April 25 - Super Energy Apocalypse, a game developed by Lars A. Doucet. The game is about teaching sustainable energy use. Abstract - Super Energy Apocalypse is my attempt at a "different" kind of educational game. The game's educational purpose is to teach basic principles about sustainable energy use. However, plenty of educational games fail at entertainment, and even at their best, the "educational" messages seem detached and unrelated to the game experience. So for Super Energy Apocalypse, the educational and entertainment parts of the game are tightly integrated: the game's setting is the fictional, post-apocalyptic republic of "New Texas" in which energy and resources are scarce and zombies are plentiful. The player has to build defenses to protect themselves from zombies, but these defenses need resources to run. These resources are energy (for flood lights and tesla coils) and metal (for gun turrets - bullets). Resources can be gained in many ways, some of which pollute more than others. Pollution causes zombies to get stronger, so by playing too aggressively and ignoring environmental concerns, the player can initiate a dangerous feedback loop. Success comes only once the player has mastered how to build up their defenses with a fairly clean, environmentally-friendly, economy. This game also took second place in a recent casual game design competition: http://jayisgames.com/archives/2008/04/super_energy_apocalypse_review.php Bio - Lars A. Doucet is a graduate student at Texas A&M University's Visualization Laboratory seeking his Master's Degree. He is very interested in video game design, for educational, artistic, and just plain entertainment purposes. Top April 18 - Screening of "Last Call for Tacos," directed by graduating Vizzer Vishwanand "V" Shetti (camera/compositing/editing) and featuring camerawork by Vizzer Chris Portales. Abstract - 'Last Call for Tacos' is a feature docudrama inspired by the life of "South Austin's Taco Queen," Maria Corbalan. As a rebellious teenager, Maria runs away from Argentina to Mexico, only to get mixed up in drugs, men, and an angel named Desiree. From abusive men in LA and drug runs in NY, Maria travels far and wide, only to find faith and stability in feeding tacos to the hungry souls of South Austin. The film is loosely based on the biography about Maria's life entited "Maridee." A work-friendly trailer can be seen at http://www.Vishwanand.com (under 'MOVIES') Bio - Vishwanand Shetti Texas A&M University - MS Visualization Science 08' Texas A&M University - BS Computer Science 03' Top April 4 - Deep Opacity Maps Abstract - This will be a practice talk of our Deep Opacity Maps paper, which I will present in EUROGRAPHICS 2008. Deep Opacity Maps is a new shadowing technique for semi-transparent objects like hair. It is extremely simple and woks in real-time with higher frame rates than the previous methods. Most importantly, it does not have the visual artifacts that are very apparent in the previous techniques. Deep Opacity Maps is also the basis of the real-time implementation of our Dual Scattering Approximation paper, which will be presented in SIGGRAPH 2008. For more info: http://www.cemyuksel.com/research/deepopacity/ Top March 28 - Exact Evaluation of Non-Polynomial Subdivision Schemes at Rational Parameter Values Abstract - We describe a method for exact evaluation of a limit mesh defined via subdivision on a uniform grid of any size. Other exact evaluation technique either restrict the grids to have subdivision sampling and are, hence, exponentially increasing in size or make assumptions about the underlying surface being piecewise polynomial (Stam's method is a widely used technique that makes this assumption). As opposed to Stam's technique, our method works for both polynomial and non-polynomial schemes. The values for this exact evaluation scheme can be computed via a simple system of linear equation derived from the scaling relations associated with the scheme or, equivalently, as the dominant left eigenvector of an up-sampled subdivision matrix associated with the scheme. To illustrate one possible application of this method, we demonstrate how to generate adaptive polygonalizations of a non-polynomial quad-based subdivision surfaces using our exact evaluation method. Our method guarantees a water-tight tessellation no matter how the surface is sampled and is quite fast. We achieve tessellation rates of over 33.5 million triangles/second using a CPU implementation. Bio - Scott Schaefer is an Assistant Professor in the Computer Science department at Texas A&M University. He graduated from Trinity University in 2000 with a B.S. degree in Computer Science and Mathematics, received an M.S. degree from Rice University in 2003 and a Ph.D. from Rice University in 2006. His research interests include Computer Graphics, Geometric Modeling and Scientific Visualization. Top February 29 - The making of a 'relatively big budget' stop-motion/live-action music video for mtvU. Abstract - Viz student Lauren Simpson recently won a contest through mtvU to direct a music video for the band Motion City Soundtrack. She enlisted fellow viz students Michael Losure and Igor Kraguljac to help create the 4 minute mixture of live-action video and stop-motion animation. The generous support of other viz lab students and staff was also crucial for the completion of the film. This talk will discuss our experiences creating the video from initial pitch all the way through to the final review process with the band, record label, and MTV. Major points of focus will include dealing with a semi-professional budget and production schedule, the technical and artistic process behind the stop-motion animation, and the interesting dynamic of juggling a band, a director, and whole lot of producer-manager-business people. Bio - Lauren, Michael, and Igor are all in their final year at the viz lab. Lauren has a B.S. in Theatre from Northwestern University. Michael has a B.S. in Computer Science and Visual Art from Union College. Igor has a BA in cinematography. Top January 23 - ImageMovers Digital Abstract - ImageMovers Digital is a new studio that is a partnership between Disney and Robert Zemeckis (director of Back to the Future 1,2,3, Who Framed Roger Rabbit, Forest Gump, Polar Express, Beowulf,...) that is recruiting heavily for R&D positions as well as Technical Directors, lighting, fx, etc... The company is entirely dedicated to "Performance Capture movies" and is currently working on "A Christmas Carol" to be released in 2009 starring Jim Carrey. Bio - Doug Epps from ImageMovers Digital would like to introduce the company, talk about what it is they do and what types of problems they have to solve. The discussion will not be technical and should be accessible to a wide audience. Top December 06 - Non Photorealistic Rendering Techniques Within a Game Framework Presented by Danny Dyer and Michael Anderson Abstract - We present several different non photorealistic rendering techniques implemented within a game framework. We will cover each of the techniques implemented and thne address how they were modified to work within the simple game we created. Bio - Mike and Danny are undergraduate students in computer science at Texas A&M. This research was done as under John Keyser. Top November 30 - Topology, Accuracy, and Quality of Isosurface Meshes Using Dynamic Particles [ Miriah Meyer, Robert M. Kirby, Ross Whitaker ] Presented by David Mayerich Abstract - This paper describes a method for constructing isosurface triangulations of sampled, volumetric, three-dimensional scalar fields. The resulting meshes consist of triangles that are of consistently high quality, making them well suited for accurate interpolation of scalar and vector-valued quantities, as required for numerous applications in visualization and numerical simulation. The proposed method does not rely on a local construction or adjustment of triangles as is done, for instance, in advancing wavefront or adaptive refinement methods. Instead, a system of dynamic particles optimally samples an implicit function such that the particles' relative positions can produce a topologically correct Delaunay triangulation. Thus, the proposed method relies on a global placement of triangle vertices. The main contributions of the paper are the integration of dynamic particles systems with surface sampling theory and PDE-based methods for controlling the local variability of particle densities, as well as detailing a practical method that accomodates Delaunay sampling requirements to generate sparse sets of points for the production of high-quality tessellations. Bio - I'm a member of the Brain Networks Lab and the Geometry and Graphics Group in the Department of Computer Science, Texas A&M. I am a PhD student and am studying under John Keyser and Bruce McCormick. Top November 16 - Real-Time Video Abstraction [Holger Winne Moller, Sven C. Olsen and Bruce Gooch] Presented by Timothy Weaver Abstract - We present an automatic, real-time video and image abstraction framework that abstracts imagery by modifying the contrast of visually important features, namely luminance and color opponency. We reduce contrast in low-contrast regions using an approximation to anisotropic diffusion, and artificially increase contrast in higher contrast regions with difference-of-Gaussian edges. The abstraction step is extensible and allows for artistic or data-driven control. Abstracted images can optionally be stylized using soft color quantization to create cartoon-like effects with good temporal coherence. Our framework design is highly parallel, allowing for a GPU-based, real-time implementation. We evaluate the effectiveness of our abstraction framework with a user-study and find that participants are faster at naming abstracted faces of known persons compared to photographs.Participants are also better at remembering abstracted images of arbitrary scenes in a memory task. Bio - Tim is a Graduate Student in Visualization at Texas A&M University (Viz Lab). His undergraduate degrees include BS in Mechanical Engineering and BFA in Studio Art. He works part time as a freelance graphic designer and as a teaching assistant for videography and design communication for the Visualization Sciences program. Currently Tim is working on the design of digital installation and performance using Max/MSP and Jitter for his graduate thesis. Top November 02 - Competition at Immersive Visualization Center About the talk - The IVC, in conjunction with the Institute for Scientific Computation, is pleased to announce its second annual Scientific Visualization Competition for Texas A&M University students. Through this competition, students will gain hands-on experience with the IVC's advanced visualization technology and will demonstrate their proficiency with scientific visualization techniques and software. (more info) Map to IVC Top October 19 - A New Graphical User Interface for a 3d Topological Mesh Modeler Abstract - In this thesis, I present a new platform-independent, open source, intuitive graphical user interface for TopMod, an application designed for interacting with 3-dimensional manifold meshes represented by a Doubly Linked Face List (DLFL). With this new interface, users will be able to create and interact with complex manifold meshes much faster and more easily than was previously possible. (more info) Bio - David Morris is a graduate student in the Visualization Sciences program at Texas A&M University. He has a Bachelor's degree in Environmental Design from Texas A&M, and currently works part-time as a teaching assistant for a 3d modeling and animation course in the Visualization Sciences program. He has also taught an undergraduate course in architectural and graphic design, computer aided drafting, and 3d modeling in the Environmental Design program and served on the Information Technology committee for the College of Architecture. Top October 12 - Presented by Lars A. Doucet [Seam Carving for Content-Aware Image Resizing by Shai Avidan, Ariel Shamir] Abstract - Effective resizing of images should not only use geometric constraints, but consider the image content as well. We present a simple image operator called seam carving that supports content-aware image resizing for both reduction and expansion. A seam is an optimal 8-connected path of pixels on a single image from top to bottom, or left to right, where optimality is defined by an image energy function. By repeatedly carving out or inserting seams in one direction we can change the aspect ratio of an image. By applying these operators in both directions we can retarget the image to a new size. The selection and order of seams protect the content of the image, as defined by the energy function. Seam carving can also be used for image content enhancement and object removal. We support various visual saliency measures for defining the energy of an image, and can also include user input to guide the process. By storing the order of seams in an image we create multi-size images, that are able to continuously change in real time to fit a given size. Link for the paper Bio (of the presenter) - Lars A. Doucet is a Master's of Visualization Sciences student at Texas A&M University. He has a Bachelor's of Environmental Design from Texas A&M. He is also founding member and Vice President of Team Sprockets, inc., a game development company based in College Station, working on a title for the XBOX 360. Top October 05 - Presented by Eric Landreneau [Pose-independent Simplification of Articulated Meshes by Christopher DeCoro, Szymon Rusinkiewicz] Abstract - Methods for triangle mesh decimation are common; however, most existing techniques operate only on static geometry. In this paper, we present a view- and pose-independent method for the automatic simplification of skeletally articulated meshes. Such meshes have associated kinematic skeletons that are used to control their deformation, with the position of each vertex influenced by a linear combination of bone transformations. Our method extends the commonly-used quadric error metric by incorporating knowledge of potential poses into a probability function. We minimize the average error of the deforming mesh over all possible configurations, weighted by the probability. This is possible by transforming the quadrics from each configuration into a common coordinate system. Our simplification algorithm runs as a preprocess, and the resulting meshes can be seamlessly integrated into existing systems. We demonstrate the effectiveness of this approach for generating highly-simplified models while preserving necessary detail in deforming regions near joints. Bio - Eric Landreneau is a Computer Science Ph.D. student at Texas A&M University. He has a B.S. and M.S. degree in Computer Science from Texas A&M. Eric currently works part-time at the Texas Center for Applied Technology as a graphics software developer. Top September 28 - Presented by Mayank Singh [Isophote Distance: A Shading Approach to Artistic Stroke Thickness ] Abstract - This paper presents an approach for determining stroke thickness in computer-generated illustrations of smooth surfaces. We assume that dark strokes are drawn to approximate the dark regions of the shaded surface. This assumption leads to a simple formula for thickness of contours and suggestive contours; this formula depends on depth, radial curvature, and light direction in a manner that reproduces aspects of thickness observed in hand-made drawings. These strokes convey local shape and depth relationships, and produce appealing imagery. Our method is simple to implement, provides temporally-coherent strokes, and runs at interactive rates. Isophote Distance: A Shading Approach to Artistic Stroke Thickness Todd Goodwin, Ian Vollick and Aaron Hertzmann | University of Toronto Bio - Top April 27 - Brian Barran [Visualization and Processing of Massive Amounts of Data] Abstract - The problem of handling and interpreting massive amounts of data is common to many industries including medical imaging, seismic interpretation, computer graphics, and more. Recent developments in parallel computing environments have provided researchers and professionals a new, simplified toolset to help grapple the task of understanding increasingly larger datasets. This presentation will show with case examples how seismic data interpretation for oil and gas exploration benefits from these new developments. In addition, a brief overview of parallel development utilizing graphics hardware will be given. Bio - Brian Barran is a graduate of the Visualization Science MS and Computer Science BS programs at Texas A&M University. He is currently Lead GPGPU Developer at Headwave, Inc -- a 3D seismic data visualization company based out of Houston -- and has worked in the past as Software Engineer at Hewlett-Packard and as a Technical Artist at Electronic Arts Canada. Top April 20 - Cem Yuksel [Deep Opacity Maps] Abstract - I will present a new method for computing shadows from semi-transparent objects like hair. Extending the concept of opacity shadow maps, our deep opacity map method uses a depth map to obtain a per pixel distribution of opacity layers. This approach totally eliminates the layering artifacts of opacity shadow maps, and requires far fewer layers to achieve high quality shadow computation. I will show qualitative comparisons to opacity shadow maps, and give performance results. The algorithm is easy to implement, fast, and memory efficient, enabling us to generate high quality hair self-shadows in real-time using consumer graphics hardware on a standard PC. Bio - Cem Yuksel received his B.S. degree in Physics and M.S. degree in Computer Engineering from Bogazici University, Turkey in 2000 and 2004 respectively. He is currently a new PhD student in Computer Science, transferred from Viz-lab MS program at Texas A&M University. His research projects include rendering, global illumination, hair modeling/simulation/rendering, and real-time water simulation. Top April 06 - Mayank Singh [A Dynamic Drawing System for Non Photorealistic Rendering] Abstract - We are exploring the idea of drawing as a time varying dynamic process. This consists of understanding how the physical entity of the pen moves over a piece of paper to create line drawings. In particular, I will be talking about a physically based control system that can be used to create drawings. The system works upon a 3D model as an input and produces a set of smooth curves. The system can be used to generate drawings that range from closely depicting the undelying 3D model to ones that have a very loosely drawn look. This work is being done in collaboration with Dr. House. Since it is still a work-in-progress, we are hoping for some lively constructive feedback. Bio - Mayank Singh received his Bachelors in Architecture from IIT, India in 2000 and his Masters in Computer Science from Marquette University in 2002. He is currently a Ph.D. student with the Department of Computer Science working with the Visualization Lab. His research interests include Non Photorealistic Techniques, Illustrative & Scientific Visualization. Top March 30 - Ergun Akleman [Interactive Modeling of Smooth Manifold Meshes with Arbitrary Topology] Abstract - In this work, we present an approach for real-time interactive topological modeling of smooth manifold meshes with arbitrary topology. Based on this approach, we have developed an interactive modeling system that allows users to work with smooth surfaces in real-time by opening or closing holes, creating handles, and combining and disconnecting surfaces. Following the tradition in parametric surface modeling, we define the shape of the smooth manifold mesh using a control mesh. Users interact and change the topological structure of a base control mesh hereinafter referred to simply as the base mesh. The smooth manifold mesh is created from the base mesh using bi-cubic patches. Each n-sided face of the base mesh is represented by n bi-cubic Bezier patches that satisfy "G1" smoothness conditions along the boundaries. We chose Bezier patches because it is simpler to specify their boundary conditions. Control vertices of the Bezier patches are automatically created based on given smoothness conditions and vertex positions of the base mesh. The paper has three main contributions: (1) We guarantee that our method will produce only smooth manifold meshes. (2) We guarantee visual smoothness for any arbitrary topological structure. (3) We demonstrate that smooth surfaces with arbitrary topology can be created interactively in real-time using parametric patches. Joint work with Vinod Srinivasan and Jianer Chen Bio - Ergun Akleman Ph.D. is an associate professor in the Department of Architecture at the Texas A&M University. He works exclusively in Master of Science in Visualization program of Architecture department. He received his M.S. and Ph.D. in Electrical and Computer Engineering from Georgia Institute of Technology jointly with Graphics, Visualization and Usability laboratory and his B.S. in Electronic engineering from Istanbul Technical University. He worked as an assistant professor in Computer Science Department of Yildiz Technical University and as an adjunct professor Computer Science Departments of Marmara and Bosphorous Universities. He is also a cartoonist who has published more than 500 cartoons, caricatures and illustrations in various newspapers and magazines. His research interests include shape modeling, artistic depiction, rendering, and visualization. Top March 23 - Scott Schaefer [Example-Based Skeletal Extraction and Skinning] Abstract - We present a method for extracting a hierarchical, rigid skeleton from a set of example poses as well as a method for estimating a set of vertex weights from those examples. We then use this skeleton to not only reproduce the example poses, but create new deformations in the same style as the examples. Since rigid skeletons are used by most 3D modeling software, this skeleton and the corresponding vertex weights can be inserted directly into existing production pipelines. To create the skeleton, we first estimate the rigid transformations of the bones using a fast, face clustering approach. We make the clustering efficient by providing a Rigid Error Function that finds the best rigid transformation from a set of points in a robust, space efficient manner and supports fast clustering operations. Next, we solve for the vertex weights and enforce locality in the resulting weight distributions. Finally, we use these weights to determine the connectivity and joint locations of the skeleton. Bio - Scott Schaefer is an Assistant Professor in the Computer Science department at Texas A&M University. He graduated from Trinity University in 2000 with a B.S. degree in Computer Science and Mathematics, received an M.S. degree from Rice University in 2003 and a Ph.D. from Rice University in 2006. His research interests include Computer Graphics, Geometric Modeling and Scientific Visualization. Top March 09 - Thesis Defense March 06, Curved Room, 10:30 A.M. Orgur Gonen [Modeling Planar Meshes with Valence-3 Vertices] - Abstract - In architectural and sculptural practice, we want to eventually construct the shapes that we have designed. In this thesis, I introduce a modeling approach to design constructible shapes. My approach guarantees that the resulting shapes are planar meshes with 3-valence vertices, which can always be physically constructed using planar or developable materials such as glass, sheet metal or plywood. March 06, Curved Room, 02:30 P.M. Can Yuksel [Real Time Impulse-based Rigid Body Simulation and Rendering] - Abstract - The purpose of this thesis is to develop and demonstrate a physically based rigid body simulation with a focus on simplifications to achieve real-time performance. This thesis aims to demonstrate that by improving the efficiency with simplified calculations of possible bottlenecks of a real-time rigid body simulation, the accuracy can be improved. A prototype simulation framework is implemented to evaluate the simplifications. The results of this prototype simulator are compared with a simulator for commercial use. Finally, various real-time rendering features are implemented not only to achieve a realistic look, but also to imitate the game-like environment where real-time rigid body simulations are mostly utilized. March 09, HRBB 302, 09:00 A.M. Brent Dingle [Volumetric Particle Modeling] - Abstract - This dissertation presents a method of modeling objects and forces for computer animation. It is an exploration of the possibilities that arise when both objects and forces are modeled using particle representations. As in most modeling systems the movement of objects is driven by physically based forces. However the usage of particles allows more artistically motivated behavior to be achieved. Such particle usage also allows the modeling of heterogeneous objects and objects in different state phases: solid, liquid, or gas. Further, by using invisible particles to propagate forces through the modeling environment it is possible to achieve complex behavior through the interaction of relatively simple components. In sum, the running theme is the emergence of 'macroscopic' behavior from 'microscopic' modeling. Top March 02 - No Brownbag Top February 23 - Cem Yuksel [Real-time Water Waves with Object Interaction] Abstract - In this talk I will present the latest results and improvements on my research with Don House and John Keyser that I presented last semester in one of our brownbag meetings. The purpose of this research is to achieve real-time performance in high quality simulation of certain water behavior with realistic object interaction. To achieve this goal we concentrate on the water surface and only simulate surface waves generated by floating object interaction. Depending on the interest of the audience I will talk about the simulation system, our mathematical formulation, and the rendering techniques I used. Bio - Cem Yuksel received his B.S. degree in Physics and M.S. degree in Computer Engineering from Bogazici University, Turkey in 2000 and 2004 respectively. He is currently a new PhD student in Computer Science, transferred from Viz-lab MS program at Texas A&M University. His research projects include rendering, global illumination, hair rendering, and real-time water simulation. Top February 16 - Ozgur Gonen [Modeling D-Forms] Abstract - This talk presents a computational method for modeling D-forms. These D-forms can directly be designed using our software. We unfold designed D-forms using a commercially available software. Unfolded pieces are later cut using a laser cutter. We obtain the physical D-forms by gluing the unfolded paper pieces together. Using this method we can obtain complicated D-forms that cannot be constructed without a computer. (The paper has been sent to Bridges'07). Bio - Ozgur Gonen received his B.A. degree in Visual Communication Design form Istanbul Bilgi University with a minor in Compuiter Science. He worked as a multimedia designer/developer for various companies in Istanbul before joining to MS in Visualization Sciences program in TAMU. He has been an active member of TopMod Developer Community since then and interested in geometric modeling. Top February 09 - Ergun Akleman [Narration Modeling] Abstract - This talk presents the concept of narration modeling and a theoretical framework to develop narration modeling systems. Similar to shape modeling systems, narration modeling systems will help users to interactively create narratives. Based on the theoretical framework, we have developed a preliminary narrative modeling system as a proof-of-concept. This preliminary narration modeling system can help users to create one panel cartoons or narrative segments in cartoon form. We have also introduced an XML file structure to represent narratives that allows users to create and change their narratives even with a text editor. Bio - Ergun Akleman Ph.D. is an associate professor in the Department of Architecture at the Texas A&M University. He works exclusively in Master of Science in Visualization program of Architecture department. He received his M.S. and Ph.D. in Electrical and Computer Engineering from Georgia Institute of Technology jointly with Graphics, Visualization and Usability laboratory and his B.S. in Electronic engineering from Istanbul Technical University. He worked as an assistant professor in Computer Science Department of Yildiz Technical University and as an adjunct professor Computer Science Departments of Marmara and Bosphorous Universities. He is also a cartoonist who has published more than 500 cartoons, caricatures and illustrations in various newspapers and magazines. His research interests include shape modeling, artistic depiction, rendering, and visualization. Top February 02 - Tracy Hammond [Sketch Recognition] Abstract - Sketching is a natural modality of human-computer interaction for a variety of tasks (e.g., conceptual design), and sketch recognition systems are currently being developed for many domains. However, they require signal-processing expertise and are time consuming to build, if they are to handle the intricacies of each domain. We want to enable user interface designers and experts in the domain itself to be able to build these systems, rather than sketch recognition experts. I created and implemented a new framework in which developers can specify a domain description indicating how domain shapes are to be identified, displayed and edited; the system then automatically generates a sketch recognition user interface for that domain. LADDER, a language using a perceptual vocabulary based on Gestalt principles and my own user studies, was developed to effectively describe how to recognize, display, and edit domain shapes. A translator and a customizable recognition system are combined with a domain description to automatically create a domain specific recognition system. With this new technology, developers will be able to write a domain description to create a new sketch interface for a domain, greatly reducing the time and expertise needed to create a new sketch interface. However, it is more natural for a user to specify a shape by drawing it than editing a text. Human perception can be used to create computer-generated descriptions from a single description. But, human and computer generated descriptions may be flawed. Thus, I created a modification of the traditional model of active learning in which the system selectively generates its own (near-miss) examples, and uses the teacher as a source of labels. System generated near-misses offer a number of advantages. Human generated examples are tedious to create and may not expose problems in the current concept. It seems most effective for the near-miss examples to be generated by whichever learning participant (teacher or student) knows better where the deficiencies lie; this will allow the concepts to be more quickly and effectively refined. When working in a closed domain such as this one, the computer learner knows exactly which conceptual uncertainties remain, and which hypotheses need to be tested and confirmed. The system uses these labeled examples to automatically build a LADDER shape description using my modification of the version spaces algorithm, which handles interrelated constraints and has the ability to also learn negative and disjunctive constraints. Bio - Tracy Hammond is an assistant professor at Texas A&M University with a focus on human perception, sketch recognition, computer human interaction, and learning. She earned the B.A. in math, the B.S. in applied math, the M.S. in computer science, the M.A. in anthropology from Columbia University and the PhD in computer science from MIT. Previously, she taught for five years at Columbia University, and she was a telecom analyst for four years at Goldman Sachs, where she designed, developed, implemented, and administers global computer telephony applications. Top January 19 - Siggraph [Updates on Siggraph 2007] Siggraph season is upon us, and everyone is madly trying to get their papers done. Relax for a bit and join us for a pizza brownbag at 12:30 Architecture C414 Curved Room. We can share our Siggraph war stories and bring everyone up to speed on what we are working on - no slides, just discussion. Eat, talk, and run back to your ball and chain (er, I mean computer). Top November 17 - Cem Yuksel [Real-time Water Waves with Two-way Object Interaction] Abstract - In this talk I will present my research with Don House and John Keyser that we are planning to submit to Siggraph 2007. This is a work-in-progress, and we are very much interested in your feedback. The method I will present simulates water waves generated by water-object interactions and simulates object motion in water. Both the simulation and rendering parts work in real-time. The simulation is scalable from a small cup of liquid to an infinite ocean, and is not limited by a predefined grid unlike grid based fluid simulations. I will talk about a new way of representing water waves, and how these waves can be generated by interacting objects. I will also present the methods to simulate the effect of water and water waves on the interacting object motion. Bio - Cem Yuksel received his B.S. degree in Physics and M.S. degree in Computer Engineering from Bogazici University, Turkey in 2000 and 2004 respectively. He is currently a new PhD student in Computer Science, transferred from Viz-lab MS program at Texas A&M University. His research projects include rendering, global illumination, hair modeling/simulation/rendering, and real-time water simulation. Top October 27 - Andrew Smith [Muscle-based Facial Animation Using Blendshapes in Superposition] Abstract - Blendshapes are an effective tool in computer facial animation, enabling represention of muscle actions. Limitations exist, however, in the level of realism attainable under conventional use of blendshapes as non-intersecting deformations. Using the principle of superposition, it is possible to create a facial model with overlapping blendshapes and achieve more realistic performance. When blendshapes overlap, the region of intersection is in superposition and usually exhibits undesired surface interference. In such cases we use a corrective blendshape to automatically remove the interference. The result is an animatable facial model implemented in Maya which represents the effects of muscle action superposition. Performance created with our model of a known human subject is compared to 3D scan reference data and video reference data of that person. Test animation is compared to video reference footage. The test animation seems to accurately mimic the effects of actual muscle action superposition. Bio - Andrew Smith received a B.S. in Computer Science from Texas A&M University in 2003 and is currently completing his M.S. in the Vizlab at Texas A&M. His area of study is facial modeling and animation techniques. Top October 20 - Zeki Melek [Visualization of Fibrous and Thread-like Data] Abstract - We will describe several methods for dealing with large amounts of thread-like data, such as data sets collected using Knife-Edge Scanning Microscopy (KESM) and Serial Block-Face Scanning Electron Microscopy (SBF-SEM). These methods allow us to collect volumetric data from embedded samples of whole-brain tissue. The neuronal and microvascular data that we acquire consists of thin, branching structures extending over very large regions. Traditional visualization schemes are not sufficient to make sense of the large, dense, complex structures encountered. We will describe interactive techniques for rendering large sets of neurons using self-orienting surfaces implemented on the GPU. We also present techniques for rendering fiber networks in a way that provides useful information about flow and orientation. Third, a global illumination framework is used to create high-quality visualizations that emphasize the underlying fiber structure. Note: This presentation is based on the paper "Visualization of Fibrous and Thread-like Data" by Z Melek, D Mayerich, C Yuksel and J Keyser, and it is going to be presented in the IEEE Viz 2006 conference in two weeks. Bio - Zeki Melek received his MS in computer science in 2000 from Bogazici University, Turkey, and is working on a Ph.D. at the Department of Computer Science here at Texas A&M University with John Keyser. His thesis is titled "Interactive Simulation of Fire, Burn and Decomposition". His research areas include physically based modeling, simulation of natural phenomena, and scientific visualization. Top October 13 - Andrew Webb, Eunyee Koh, Andruid Kerne [Choreographic Buttons: Promoting Social Interaction through Human Movement and Clear Affordances] Abstract - We present an interactive installation that introduces a new type of affordance, the choreographic button, which integrates movement, gesture recognition, and interactive visual feedback. Our intention was to promote social interaction and creative emergence. Jumping, a quick movement, and crouching, a sustained gesture, were choreographed to form a vocabulary that is personally expressive and which also facilitates automatic recognition. Arranged in a 3 x 3 grid in the installation, these choreographic buttons are used to explore a collection of consonant imagery, which consists of personal movements of extension and contraction. The movements function as expressions through which participants dynamically create a series of individual and collaborative compositions. Narrative and lyrical components emerge through elements and patterns. Movements in the installation contribute to an embodied process of interaction that integrates dance performance with control structure and feedback, engaging onlookers in each successive participant's creative process. By conducting an integrated user study / art opening event, we found this promotes social interaction between participants and onlookers. For more information, please visit -- [http://ecologylab.cs.tamu.edu/cb/] Bio - Dr Andruid Kerne - I am a research scientist-artist investigating how people experience personal expression, creative ideation, and social engagement. I develop and evaluate expressive interfaces, computational architectures, and distributed systems that support creative processes of knowledge production and interpersonal communication. Essential research areas are human computer interaction, knowledge management, digital libraries, information visualization, gaming, wearable physiological computing, sensor networks, pervasive computing, embedded systems, and software engineering. Intersecting fields include pattern recognition, information retrieval, cognitive psychology, visual and interaction design, time-based media, ethnography, performance studies, music composition, and conceptual art. http://ecologylab.cs.tamu.edu/index.html For more information about Dr. Kerne and his group, please visit -- [http://ecologylab.cs.tamu.edu/index.html] Top October 06 - Alethea Bair [Texturing of Layered Surfaces for Optimal Viewing] Abstract - We take a new look at the problem of texturing surfaces so that they can be displayed layered over each other but remain clearly visible. Finding optimal textures that solve this problem is complex because of the perceptual interactions between the visual effects of parameters controlling texture generation. Instead of using controlled experiments to investigate this problem, we use a genetic algorithm based human-in-the-loop parameter space search to build a large database of human-rated textures. This database is then analyzed with a variety of data- mining techniques, including ANOVA, linear discrimant analysis, decision trees, parallel component analysis. We detail this analysis, concluding with a set of guidelines for building strong layered surface textures, and a display of a number of example textures. Bio - Alethea Bair received a degree in physics from the University of Illinois Urbana/Champaign in 2003, and is currently working on a Ph.D. at the Vizlab here at Texas A&M University. Her area of study is scientific visualization. Top September 29 - Charles Loop [Direct GPU Rendering of Higher Order Shapes] Abstract - Driven by the lucrative video game industry's quest for greater realism, the Graphics Processing Unit has evolved into a programmable SIMD machine. The stages of this hardware pipeline are specialized to accelerate vertex processing and triangle rasterization. These two geometric operations are designed specifically for rendering densely triangulated surfaces. In this talk, I will discuss how to use the GPU programming model to render curves and surfaces directly in terms of their defining equations, rather than as a collection of approximating linear elements. This approach uses far less memory and produces real-time images that are resolution-independent. Bio - Charles Loop is a Researcher with Microsoft Research in Redmond, Washington. He has a Ph.D. in Computer Science from the University of Washington, 1992. He has published papers on Curve and Surface Modeling, Computer Vision techniques, and GPU algorithms. Top September 22 - Robert Harriss & William Merrell [Visualization Problems and Opportunities for Coastal Environments] Abstract - Bob Harriss and Bill Merrell will share with us projects that they are involved with or are developing at HARC and TAMU-G that contain interesting opportunities for visualization application, and interesting challenges for visualization research. This will be an informal presentation, that should be highly interactive. Come prepared for an interesting discussion. Bio - Robert Harriss, Houston Advanced Research Center (HARC) William Merrell, Texas A&M Galveston Top September 15 - Jinxiang Chai [Interative Control for Realistic Character Animation] Abstract - Feature films exhibit stunningly realistic digital actors and video games contain responsive and controllable virtual characters. However, currently available animation technologies are not useful tools for naive users because they either require a tremendous amount of artistry, skill, and time, or rely on commercial motion capture devices that are invasive and expensive. This talk focuses on three intuitive interfaces that allow the naive user to easily and quickly generate realistic character animation. Each solution relies on the information about natural human motion inherent in a motion capture database. First, I introduce a performance animation that uses video input of the user to build a local model of the user's motion and reproduce it on an animated character. The second interface allows the user to control realistic facial expression by acting out a desired motion in front of a single video camera. And finally, I have developed a constraint-based motion optimization technique for generating natural-looking facial and full-body character animation from a small set of user-defined constraints. We assess the quality of the resulting animation by comparisons with those created by a commercial optical motion capture system. Bio - Jinxiang Chai joined the Texas A&M University in fall 2006 as an assistant professor of computer science. He received his Ph.D from Carnegie Mellon's School of Computer Science. His research interests span computer animation, computer graphics, and computer vision. He is particularly interested in developing methods and systems that allow everyone to create and manipulate high-dimensional visual data such as character animation, 3D geometic models, images, and videos as easy as text. He is also interested in interactions techniques for 3D graphics, data-driven approach for graphics and vision, computer vision techniques for graphics and animation applications, image-based rendering and modeling, image and video processing. Top September 01 - Scott Schaefer [Approximate Catmull-Clark Patches] Abstract - We present a simple and computationally efficient algorithm for approximating Catmull-Clark subdivision surfaces using a minimal set of bicubic patches. For each quadrilateral face of the control mesh, we construct a geometry patch and a pair of tangent patches. The geometry patches approximate the shape and silhouette of the Catmull-Clark surface and are smooth everywhere except along patch edges containing an extraordinary vertex where the patches are C^0. To make the patch surface appear smooth, we provide a pair of tangent patches, which approximate the tangent fields of the Catmull-Clark surface. These tangent patches are used to construct a continuous normal field for shading and displacement mapping. Using this bifurcated representation, we are able to define an accurate proxy for Catmull-Clark surfaces that is efficient to evaluate on next-generation GPU architectures that expose a programmable tessellation unit. Bio - Scott Schaefer is an Assistant Professor in the Computer Science department at Texas A&M University. He graduated from Trinity University in 2000 with a B.S. degree in Computer Science and Mathematics, received an M.S. degree from Rice University in 2003 and a Ph.D. from Rice University in 2006. His research interests include Computer Graphics, Geometric Modeling and Scientific Visualization. Top April 28 - Unmil Karadkar & Marlo Nordt [Image-based Evaluation of Video-acquired Research Skills] Abstract - Digital image collection interface layouts vary in the nature and degree of contextual information they provide to their users thus enabling or impeding specific tasks. We are exploring image presentation techniques to support image-centric cognitive tasks in the context of cardiovascular systems research and education. To investigate the effect of image layout on user performance, we conducted an experimental evaluation of three image layouts for three representative tasks in this domain. The layouts varied the spatial and temporal presentation of images thus providing different contextual information to the test subjects. Our results indicate that the degree of contextual information provided by the image layouts affected user performance, as did their research expertise. These results will inform the design of user interfaces for performing image-focused cognitive tasks as well as the development of interfaces for training novice researchers. Bio - Unmil Karadkar - is a Ph.D. candidate in the Department of Computer Science at Texas A&M University. His work focuses on human-centered design and experimental evaluation of information systems for individuals and groups in different disciplines. His research interests include Human-Computer Interaction, Digital Libraries, Hypermedia Systems, Educational Technology, and Information Retrieval. He holds a M.S. in Computer Science from Texas A&M University and a B.E. in Computer Engineering from University of Pune, India. Marlo Nordt - is also a Ph.D. candidate in the Department of Computer Science at Texas A&M University. Her area of focus is Human Computer Interaction (HCI). She graduated with my B.S. in Computer Science in May 2003 from Stephen F. Austin State University (SFA) in Nacogdoches, TX. During her time at SFA, she played volleyball for the school which is a member of the Division I Southland Conference. Top April 21 - Dr. Yongkui Liu [Several Fundamental Algorithms In Computer Graphics] Abstract - In this talk, some published works by Dr. Yongkui Liu will be briefly introduced. It includes a method to approximate the circle, an algorithm for Boolean operations between regions of polygons, an algorithm for voxel traversing along a straight line and a compressed chain code. Graphics display on the hexagonal grid is also discussed. Bio - Professor in Computer Science College, Dalian Nationalities University Top March 31 - Burak Pak [Design Computing Studio : Paradigms, Processes and Future Projections] Abstract - The presentation will focus on the use of design computing methods and approaches in design process through various strategies and models. Different cases produced in design computing studios at numerous universities (including the workshops and studios coordinated by the presenter) will be reviewed to discuss the subject. Future projections on implementation of digital environments and the use of computational tools for design and education will also be covered. Bio - B. Sc in Architecture (Yildiz Technical University) M. Sc in Architectural Design Computing (Istanbul Technical University) Pursuing a PhD in Istanbul Technical University, Faculty of Architecture Research interests: Digital design and education, design computing, generative design, design technologies and virtual environments. Top March 03 - David Eberle & Paul Edmonson [Tools in the Visual Media] Abstract - Whether it's for feature films or video games, tools are an essential part of visual media production. Join two industry developers as they talk about the differences between writing code and tools in academia, games, and movie production houses. They will also talk about common misconceptions about industry development and how to break into what seem to be very exclusive fields. Production shots and from-the-field software demonstrations will also be shown. Bio - Dave Eberle graduated from the Math Department at A&M with an emphasis in computer graphics and simulation. He started at Disney Feature Animation working on the dynamics of the skin system for the movie "Reign of Fire." After Disney he worked for the physics engine middleware company Havok which specializes in realtime rigid body dynamics. He now works at PDI/Dreamworks in the FX tools division of R&D where he is currently focusing on deformable simulation domains. Paul Edmondson has been working as a Technical Art Director and Tool Developer in the video game industry since he finished his degree in Computer Science at A&M. He has been involved in the development of the Star Wars: Knights of the Old Republic game series (LucasArts), Lord of the Rings: Battle for Middle Earth (Electronic Arts), and America's Army: Rise of a Soldier (Secret Level). Paul is currently at Secret Level in San Francisco, working as a Tool Architect on a next-generation console game for Sega. Top February 24 - Ricardo Gutierrez-Osuna [Audio-visual mappings for speech-driven facial animation] Abstract - Visual information plays a significant role in face-to-face spoken communication. Facial motion is essential to the hearing-impaired, and is also used by normal listeners as a supplement to improve speech intelligibility in noisy conditions. Advances in speech synthesis and computer graphics during the past two decades have allowed computer facial animation to become a prime tool for research in cross-modal speech perception. The long-term goal of our research is to create highly expressive facial animations that can be used as assistive and educational technology for speech perception in persons with communication disabilities. Our vision is a new generation of real-time facial animations capable of articulating not only lip movements but also visual prosody (e.g., head nods, eyebrow raises) directly from speech acoustics i.e., without a speech transcript. The specific objective of this talk is to introduce computational methods for learning audio-visual mappings from data. We will review the various techniques that can be used to animate facial models, present our previous results on the synthesis of lip movements from sub-phonemic acoustics, and discuss our current research efforts on modeling visual prosody. Bio - Ricardo Gutierrez-Osuna received the B.S. degree in Electrical Engineering from the Polytechnic University of Madrid (Spain) in 1992, and the M.S. and Ph.D. degrees in Computer Engineering from North Carolina State University in 1995 and 1998, respectively. From 1998 to 2002 he served on the faculty at Wright State University. He is currently an assistant professor in the Department of Computer Science at Texas A&M University. Dr. Gutierrez-Osuna is a recipient of the National Science Foundation Career Award for his research on machine olfaction with chemical sensors arrays. His research interests include pattern recognition, biological cybernetics, sensor instrumentation, speech-driven facial animation, and mobile robotics. Top February 17 - Cem Yuksel [Hair Occlusion: Rendering Hair with Global Illumination] Abstract - I will be talking about hair occlusion, a general framework that is particularly suitable for global illumination of hair-like long, thin, and semi-transparent objects that can be represented by a series of line segments. This framework is compatible with Monte Carlo ray-tracing, and it can be integrated into the final gathering stage of photon mapping. Image based illumination techniques are also supported. Using this framework we can render hair images with various realistic lighting simulations that are not supported by previous hair rendering methods. I will also talk about optimization techniques to reduce the computational complexity, allowing several orders of magnitude faster hair occlusion calculations. Bio - BS in Physics, Bogazici University, Turkey (1995-2000) MS in Computer Engineering, Bogazici University, Turkey (2000-2003) Pursuing MS in Visualization Sciences, Texas A&M University (2004-2006) Research Interests: Rendering & global illumination, physically based simulation Top February 10 - Brent Dingle [All Things Particle, Take 1] Abstract - I will briefly describe a modeling paradigm where every object is volumetrically modeled as a composition of particles. From this I will discuss the possibility of also representing forces as particles. In this representation the motion of the forces is modeled as moving force particles which may act on other particles. This allows for complex force behavior and interaction to be modeled and implemented in a straightforward fashion. A demonstration of some of the visual effects which can be achieved using this concept will be presented. Top February 3 - Alex Timchenko [Image Correction for Immersive Visualization Systems] Abstract - In this talk, I will discuss the issues present in generating spatially coherent images for multifaceted immersive displays. The use of commodity projectors can lower the cost of immersive visualization systems; however, such projectors are difficult to align and introduce perspective and barrel distortion. The talk will cover these problems and the various methods and interfaces for dealing with them, as well as possible future solutions. Top December 2 - Fred Parke [Next Generation Spatially Immersive Visualization Systems] Abstract - This presentation focuses on the development of lower cost, modular immersive visualization systems. Technology now available enables spatially immersive visualization systems created using off the shelf components including high performance, relatively inexpensive, commodity computers, inexpensive commodity projectors and open source software. Modular configurations utilizing polyhedral display surfaces with many identical modular components and networked visual computer clusters is one approach to such systems. Previous work at the Texas A&M College of Architecture, funded by a TAMU TITF grant, focused on developing and evaluating several prototypes of this new class of systems to determine their practicality and effectiveness. Current work is focused on implementing two production quality systems funded by an NSF MRI development grant. Underlying concepts, issues and trade-offs related to the design and development of these systems will be presented. Top November 18 - Ergun Akleman [Intriguing Relation between Geometry and Mesh Topology and Its Effects on Subdivision Modeling] Abstract - In this paper, we introduce the concept of vertex angle deviation for piecewise planar manifold meshes. Vertex angle deviation gives similar or better information about local behavior than Gaussian curvature. Moreover, if we add angle deviations of all vertices the result turned out to be independent of mesh structure and it depends only the genus of the surface. Based on this result, it can be possible to improve organization the mesh structure of a shape according to its geometric structure. Top November 4 - John Keyser [Robust solid modeling: history, progress, and future directions] Abstract - Problems with robustness, related to both numerical instability and degeneracies, have long plagued solid modeling applications. In this talk, I will review sources of these robustness problems, the history of research on these topics, the recent research both by my group and others, and some of the major open issues/problems remaining (including thesis/dissertation directions). The talk will be at an overview level, without assuming much mathematical background, but rather just a very basic familiarity with geometric modeling. Top October 28 - Zeki Melek [Pyromania, or "How I stopped burning stuff and started simulating"] Abstract - I will present a modular simulation framework to integrate several aspects of the combustion and burning process in a unified manner. Combustion reaction, hot air motion, objects catching fire, starting to burn and decompose all in [barely] interactive simulation. I will discuss models used for each phenomea, as well as some of the work in progress modules, including simulation control and curling of burning matches. Top October 21 - Jacob Beaudoin [Large-scale real-time vegetation simulation and rendering] Abstract - I will talk about simulations on massive scales and how simulation level of detail techniques as well as modern and future graphics hardware help make it possible. This includes some of the work we have done in simulation level of detail for trees and grass as well as some work in progress for enhancing these simulations and rendering them in real-time. Top October 14 - Jyh-Ming Lien [Approximations in shape and motion computations] Abstract - Shape and motion are essential in many areas of geometric computations. We have developed frameworks that allow us to approximate shape and motion in geometric computations. Our study of a wide range of applications shows that approximations of shape and motion can not only provide computational efficiency but can also allow multi-resolution or hierarchical representations. More importantly, an approximate representation can provide a way to ignore insignificant details and enable Us to concentrate on features that are more significant. This gives us the ability to comprehend a problem from a global perspective when a closer inspection is not possible. This talk will provide an overview of our frameworks and their applications. We will first describe a general shape approximation method, named approximate convex decomposition, and then describe a more specific approximate method for cortical network reconstruction. Our focus of approximating motion is on generating motion using motion-planning methods. In addition, our study shows evidence that approximate shape (motion) can provide a better understanding of and more insight into an object's motion (shape) than exact shape and motion representations. We will use the problem of motion planning of deformable object as an example to illustrate this observation. Finally, we will describe how we can simulate complex group behaviors by allowing the agents to have limited access to an approximated representation of the reachable space in an environment. Top October 7 - Barani Raman [Sensor based machine olfaction with neuromorphic models of the olfactory system] Abstract - In this talk, we present a biologically inspired architecture for pattern recognition for chemosensor array inspired by key signal processing mechanisms in the olfactory system. Our approach can be summarized as follows. First, a high-dimensional odor signal is generated from a chemical sensor array. Three approaches have been proposed to generate a combinatorial and high dimensional odor signal: temperature-modulation of metal-oxide chemoresistor, a large population of optical microbead sensors, and infrared spectroscopy. The resulting high-dimensional odor signals are subject to dimensionality reduction using a self-organizing model of chemotopic convergence. This convergence transforms the initial combinatorial high-dimensional code into an organized spatial pattern (i.e., an odor image), which decouples odor identity from intensity. Two lateral inhibitory circuits subsequently process the highly overlapping odor images obtained after convergence. The first shunting lateral inhibition circuits perform gain control enabling identification of the odorant across a wide range of concentration. This is followed by an additive lateral inhibition circuit with center-surround connections. These circuits improve contrast between odor images leading to more sparse and orthogonal patterns than one available at the input. The sharpened odor image is stored in a neurodynamic model of cortex. Finally, anti-Hebbian/ Hebbian inhibitory feedback from the cortical circuits to the contrast enhancement circuits performs mixture segmentation and weaker odor/background suppression respectively. We validate the models using experimental datasets and show our results are consistent with recent neurobiological findings. Top September 30 - David Mayerich [A New Method for the Acquisition and Reconstruction of High-Resolution Tissue Data Sets] Abstract - The reconstruction and simulation of networks in the mammalian brain requires detailed anatomical models that describe how cells are distributed and connected. Standard stochastic models work well for the simulation of individual cells, but they lack the connectivity information necessary to create large-scale networks. The Knife-Edge Scanning Microscope (KESM) was designed with the goal of gathering large scale 3D datasets with the resolution necessary to reconstruct mammalian brain networks. I will give a short overview of the methods we are using to gather large scale datasets and provide current examples of the types of data that can be extracted from these datasets, including vascular networks, cell distribution, and the morphology of individual neurons. Top September 9 - Wei Yan [Simulating Users' Behavior in Built Environments] Abstract - This talk will describe a behavior simulation model that addresses the problem of predicting and evaluating the impacts of the built environment on its human inhabitants, which is one of the most important, yet difficult performances to predict and evaluate before the environment has actually been con-structed. We have developed a computer simulation system that simulates the environment “in action”, much like electrical and mechanical engineers can “run” their designs to see how they perform under certain conditions. The simulation consists of a usability-based building model and an agent-based virtual user model. The building model, unlike traditional CAD models, possesses both graphical/geometric information of design elements and non-graphical information about usability properties of these elements. We have developed a method to automatically convert standard CAD models to usability-based models, whose environmental information is structured in a way that makes it perceivable and interpretable by the virtual users. The virtual users are modeled as autonomous agents that emulate the appearance, perception, social traits and physical behavior of real users. User modeling consists of geometry modeling, perception modeling, and behavior modeling. Behavior modeling is the most critical issue underlying the simulation because it must mimic closely how humans behave in similar socio/spatial environments, given similar goals. Accordingly, our behavior modeling stems from three important and firm sources: 1) Theoretical and practical environment-behavior studies that provided us with the basic function of environment-behavior relationship. 2) A field study assisted by a video tracking system using computer vision techniques, which provided substantial statistical measurements about users’ behavior. 3) Artificial Life (ALife) research, which provided primitive group behavior algorithms for simulating spatial interactions among individuals during their movements. Built upon these three sources, the virtual users can exhibit similar traits to those observed in reality. By inserting these virtual users in the usability-based building model and letting them “explore” it on their own volition, the system reveals the interrelationship between the environment and its users. We expect the result of this research to change how architects and environmental behavior experts will approach the design and evaluation of built environments. Top |