Graphics Lunch

Home Members Applications Projects Seminars Publications Facilities Internal Web Contact Us

Search

Graphics Lunch - Spring, 2007

Graphics Lunch is a forum for informal and formal discussions over lunch for those interested in graphics and visualization issues at Maryland. It also serves as a forum for talks from visitors to our lab about their recent research in graphics and visualization. Students and faculty can use this venue to practise and prepare for their conference papers, discuss recent and upcoming papers and conferences, or inform others about graphics and visualization news. Meetings are held on Mondays from 12:00pm to 1:30pm in AVW 4185 .
Announcements regarding Graphics Lunch will be sent to the graphics-local mailing list. You may join this list at http://www.cs.umd.edu/mailman/listinfo/graphics-local.

Recent Seminars

April 16, 2007 Sound Technology in Games

Presented By Kenny Weiss, University of Maryland

Comments Graphics Seminar Series
Slides from this talk

Abstract
We discuss issues related to sound in games including audio hardware, APIs and the aural rendering pipeline. Audio games (games with an audio output rather than a visual one) and the various roles for sound designers in gaming will also be discussed.
Mar 26, 2007 Modeling Non-manifold simplicial three-dimensional shapes

Presented By Annie Hui, University of Maryland

Comments Graphics Seminar Series

Abstract
In the field of solid modeling, research has focused predominantly on manifold shapes because they are ``smooth''. For example, terrains are often modeled as layers of tetrahedras. Most models in the Princeton Shape Benchmark are made up of manifold surfaces describing the skin of the objects modeled. However, the class of manifold shapes makes up only a small category of solid shapes. The remaining, larger category of shapes is generally non-manifold. Historically, this category has been evaded because of the irregularity of non-manifold shapes.

In my work, I review existing approaches that tackle non-manifold shapes. I discuss the topological properties of non-manifold shapes, and present my proposed approaches to non-manifold shape modeling, which optimize on the time and space use. These representations are the basis on which applications are built to better shape modeling.

Upon this basis, I designed tools for manipulating and understanding non-manifold shapes. The first tool is the elementary mesh modification operators I proposed for the non-manifold representations. Two applications which benefit from this tool are the idealization of CAD models and multi-resolution modeling in Computer Graphics. Another tool is the decomposition of non-manifold shapes. An object can be decomposied based on connectivity and dimension. The connectivity of the resultant components in decomposition can be reprsented as a graph which describes the structure of the object. This information important to applications on shape characterization and understanding. It is a useful tool for the management of WEB-based shape retrieval.
Mar 5, 2007 Volumetric Contour-Based Surface Reconstruction

Presented By David Breen, Drexel University

Comments Graphics Seminar Series
Held in AVW 3258

Abstract
This talk will present three techniques for the creation of closed, smooth surfaces from a set of noisy parallel contours sampled at a range of resolutions. Imaging technology, e.g. MRI, CT, ET and histology, is now widely used in medicine, science and engineering to study the internal structures of a variety of objects and specimens. This technology produces 3D sampled data that can be interpreted as a stack of 2D slices cutting through the studied object/specimen. Frequently the process of isolating, segmenting and identifying specific structures in the slices involves a manual (or semi-automatic at best) process of delineation that produces contours around the structure of interest in each slice. The different imaging technologies and delineations produce contour sets with varying noise properties and sampling resolutions. A single reconstruction technique is therefore unlikely to produce satisfactory results for all types of contour input.

In order to solve this problem for many contour input types an effort is underway to develop and study three volumetric contour-based surface reconstruction techniques. A volumetric/implicit approach has been taken because these types of techniques easily handle changes in topology and more robustly reconstruct complex multi-component, branching objects. The first two techniques are more suitable for sparse contour sets, i.e. those from a low number of high resolution slices, and are based on 2D level set morphing and 2D distance field interpolation. The third method is more appropriate for high density input that contains significant noise both in-plane and inter-plane, and utilizes an implicit point set model to create a smooth surface with user-specified error bounds. The details and comparisons of these three methods will be presented, along with numerous reconstruction results from a variety of contour datasets.
Feb 26, 2007 Active Illumination based 3D Surface Reconstruction and Registration for Image Guided Medialization Laryngoplasty

Presented By Jin Ge, The George Washington University

Comments Graphics Seminar Series

Abstract
The medialization laryngoplasty is a surgical procedure to improve the voice function of the patient with vocal fold paresis and paralysis. An image guided system for the medialization laryngoplasty will help the surgeons to accurately place the implant and thus reduce the failure rates of the surgery. One of the fundamental challenges in image guided system is to accurately register the preoperative radiological data to the intraoperative anatomical structure of the patient. In this talk, I will present a combined surface and fiducial based registration method to register the preoperative 3D CT data to the intraoperative surface of larynx. To accurately model the exposed surface area, a structured light based stereo vision technique is used for the surface reconstruction. We combined the gray code pattern and multi-line shifting to generate the intraoperative surface of the larynx. To register the point clouds from the intraoperative stage to the preoperative 3D CT data, a shape priori based ICP method is proposed to quickly register the two surfaces. The proposed approach is capable of tracking the fiducial markers and reconstructing the surface of larynx with no damage to the anatomical structure. We used off-the-shelf digital cameras, LCD projector and rapid 3D prototyper to develop our experimental system. The final RMS error in the registration is less than 1mm.
Feb 5, 2007 Distinctive Regions of 3D Surfaces

Presented By Phil Shilane, Princeton University

Comments Graphics Seminar Series
Held in AVW 3258

Abstract
Selecting the most ``important'' regions of a surface is useful for a variety of applications in computer graphics and geometric modeling. While previous research has analyzed geometric properties of meshes in isolation, we select regions that distinguish a shape from objects of a different type. Our approach to analyzing distinctive regions is based on performing a shape-based search using each region as a query into a database. Distinctive regions of a surface have shape consistent with objects of the same type and different from objects of other types. For a classified database of models, distinction can be analyzed in a pre-processing phase, but we also use a likelihood model to predict distinction for new shapes based on a training set. We demonstrate the utility of detecting distinctive surface regions in several applications, including mesh visualization, shape matching, icon generation, and mesh simplification.

Old Seminars

2007(Spring/Fall)

2006(Spring/Fall)

2005

2004

2003

2002(Summer/Spring/Fall)

2001

Top

April 16, 2007	Sound Technology in Games
Presented By	Kenny Weiss, University of Maryland
Comments	Graphics Seminar Series Slides from this talk
Abstract	We discuss issues related to sound in games including audio hardware, APIs and the aural rendering pipeline. Audio games (games with an audio output rather than a visual one) and the various roles for sound designers in gaming will also be discussed.
Mar 26, 2007	Modeling Non-manifold simplicial three-dimensional shapes
Presented By	Annie Hui, University of Maryland
Comments	Graphics Seminar Series
Abstract	In the field of solid modeling, research has focused predominantly on manifold shapes because they are ``smooth''. For example, terrains are often modeled as layers of tetrahedras. Most models in the Princeton Shape Benchmark are made up of manifold surfaces describing the skin of the objects modeled. However, the class of manifold shapes makes up only a small category of solid shapes. The remaining, larger category of shapes is generally non-manifold. Historically, this category has been evaded because of the irregularity of non-manifold shapes. In my work, I review existing approaches that tackle non-manifold shapes. I discuss the topological properties of non-manifold shapes, and present my proposed approaches to non-manifold shape modeling, which optimize on the time and space use. These representations are the basis on which applications are built to better shape modeling. Upon this basis, I designed tools for manipulating and understanding non-manifold shapes. The first tool is the elementary mesh modification operators I proposed for the non-manifold representations. Two applications which benefit from this tool are the idealization of CAD models and multi-resolution modeling in Computer Graphics. Another tool is the decomposition of non-manifold shapes. An object can be decomposied based on connectivity and dimension. The connectivity of the resultant components in decomposition can be reprsented as a graph which describes the structure of the object. This information important to applications on shape characterization and understanding. It is a useful tool for the management of WEB-based shape retrieval.
Mar 5, 2007	Volumetric Contour-Based Surface Reconstruction
Presented By	David Breen, Drexel University
Comments	Graphics Seminar Series Held in AVW 3258
Abstract	This talk will present three techniques for the creation of closed, smooth surfaces from a set of noisy parallel contours sampled at a range of resolutions. Imaging technology, e.g. MRI, CT, ET and histology, is now widely used in medicine, science and engineering to study the internal structures of a variety of objects and specimens. This technology produces 3D sampled data that can be interpreted as a stack of 2D slices cutting through the studied object/specimen. Frequently the process of isolating, segmenting and identifying specific structures in the slices involves a manual (or semi-automatic at best) process of delineation that produces contours around the structure of interest in each slice. The different imaging technologies and delineations produce contour sets with varying noise properties and sampling resolutions. A single reconstruction technique is therefore unlikely to produce satisfactory results for all types of contour input. In order to solve this problem for many contour input types an effort is underway to develop and study three volumetric contour-based surface reconstruction techniques. A volumetric/implicit approach has been taken because these types of techniques easily handle changes in topology and more robustly reconstruct complex multi-component, branching objects. The first two techniques are more suitable for sparse contour sets, i.e. those from a low number of high resolution slices, and are based on 2D level set morphing and 2D distance field interpolation. The third method is more appropriate for high density input that contains significant noise both in-plane and inter-plane, and utilizes an implicit point set model to create a smooth surface with user-specified error bounds. The details and comparisons of these three methods will be presented, along with numerous reconstruction results from a variety of contour datasets.
Feb 26, 2007	Active Illumination based 3D Surface Reconstruction and Registration for Image Guided Medialization Laryngoplasty
Presented By	Jin Ge, The George Washington University
Comments	Graphics Seminar Series
Abstract	The medialization laryngoplasty is a surgical procedure to improve the voice function of the patient with vocal fold paresis and paralysis. An image guided system for the medialization laryngoplasty will help the surgeons to accurately place the implant and thus reduce the failure rates of the surgery. One of the fundamental challenges in image guided system is to accurately register the preoperative radiological data to the intraoperative anatomical structure of the patient. In this talk, I will present a combined surface and fiducial based registration method to register the preoperative 3D CT data to the intraoperative surface of larynx. To accurately model the exposed surface area, a structured light based stereo vision technique is used for the surface reconstruction. We combined the gray code pattern and multi-line shifting to generate the intraoperative surface of the larynx. To register the point clouds from the intraoperative stage to the preoperative 3D CT data, a shape priori based ICP method is proposed to quickly register the two surfaces. The proposed approach is capable of tracking the fiducial markers and reconstructing the surface of larynx with no damage to the anatomical structure. We used off-the-shelf digital cameras, LCD projector and rapid 3D prototyper to develop our experimental system. The final RMS error in the registration is less than 1mm.
Feb 5, 2007	Distinctive Regions of 3D Surfaces
Presented By	Phil Shilane, Princeton University
Comments	Graphics Seminar Series Held in AVW 3258
Abstract	Selecting the most ``important'' regions of a surface is useful for a variety of applications in computer graphics and geometric modeling. While previous research has analyzed geometric properties of meshes in isolation, we select regions that distinguish a shape from objects of a different type. Our approach to analyzing distinctive regions is based on performing a shape-based search using each region as a query into a database. Distinctive regions of a surface have shape consistent with objects of the same type and different from objects of other types. For a classified database of models, distinction can be analyzed in a pre-processing phase, but we also use a likelihood model to predict distinction for new shapes based on a training set. We demonstrate the utility of detecting distinctive surface regions in several applications, including mesh visualization, shape matching, icon generation, and mesh simplification.