Engineering Departments

Computer Science


From left: Bert Huang, Kurt Luther, and Sharath Raghvendra
Blacksburg, VA , September 03, 2014
Virginia Tech College of Engineering

Bert Huang

Research Interests:

His name is on six patents, ranging from machine learning for power grids to ways to analyze spatiotemporally ambiguous events to combinatorial optimization methods and systems. Since fall of 2011, Huang has worked as a postdoctoral research associate at the University of Maryland’s Department of Computer Science. He conducts collaborative research on machine learning in network and relational domains, focusing on topics including large-scale probabilistic methods and computational learning theory for structured models.

Education:

Master's Degree: Philosophy in Computer Science, Columbia University

Master's Degree: Computer Science, Columbia University

Doctoral Degree: Computer Science, Columbia University
 

Kurt Luther

Research Interests:

Luther’s main research interests include human–computer interaction (HCI), social computing and crowdsourcing, and creativity support tools. Specifically, building and studying social computing systems to enhance human creativity and problem solving abilities, in domains such as computer animation, visual design, knowledge discovery, and citizen science. Also interested in connections to the digital humanities, especially history.

Education:

Bachelor's Degree: Computer Graphics Technology, Minor in Art and Design, Purdue University

Doctoral Degree: Human-Centered Computing, Georgia Tech
 

Sharath Raghvendra

Research Interests:

Raghvendra focuses on the design of algorithms for geo-metric problems. He is particularly interested in the creation of algorithmic tools and methodologies which are applicable to large-scale, unstructured, and potentially very high dimensional geometric data.

Education:

Bachelor's Degree: Computer Science and Engineering, Indian Institute of Information Technology (IIIT), Hyderabad, India

Doctoral Degree: Computer Science, Duke University



Allison Tegge
Blacksburg, VA , August 28, 2014
Virginia Tech College of Engineering

Allison Tegge, a post-doctoral associate researcher with Virginia Tech’s Department of Computer Science, has been awarded a three-year National Institutes of Health fellowship worth $150,000.

Tegge was awarded the Individual Postdoctoral Fellowship, also known as an F32, by the National Institute of Environmental Health Sciences through the Ruth L. Kirschstein National Research Service Awards program. The fellowships provide up to three years of support for promising postdoctoral researchers who have the potential to become productive, independent investigators within the broad scope of biomedical, behavioral, or clinical research, according to the agency’s website.

The fellowship is tied to Tegge’s work on researching how liver cells signal one another. The process is not yet understood and is a focal point of research by Virginia Tech College of Engineering faculty Padma Rajagopalan, associate professor of chemical engineering, and T. M. Murali, associate professor of computer science. Murali and Rajagopalan will co-mentor Tegge on her research.

Rajagopalan has developed an in vitro 3-D bioengineered liver tissue that contains three major cell types. In humans, the liver is primarily responsible for metabolizing foreign compounds such as alcohol, cigarette smoke, and drugs. Rajagopalan and Murali are collaborating on discovering how these cell types may be communicating with each other in order to maintain cellular function.

Their combined team is working on the effort as part of the Virginia Tech Institute for Critical Technology and Applied SciencesCenter for Systems Biology of Engineered Tissues, with funding from the National Science Foundation and the Environmental Protection Agency. Tegge’s work focuses on the computational aspects of the research. She is developing computer algorithms that in essence will discover the code for inter-cellular signaling within the liver, which, in turn, could lead to a similar understanding of other body organs.

“Allison’s research is at the interface of computer science and tissue engineering, and it will map how protein pathways that mediate inter-cellular signaling may be perturbed by exposure to environmental chemicals,” said Murali. “She will develop cutting-edge graph-theoretic algorithms to discover these patterns.”

Part of the in-lab research by Rajagopalan involves exposing the liver tissue models to various toxic environmental chemicals. Researchers then watch and test how the cells within the sample react to the administration, including activation of protein signaling pathways. Tegge’s algorithms will help “discover” which protein pathways create a signal that can cause observed changes in gene expression, added Murali.

“This award from the National Institutes of Health is a testament to the promise of her research,” said Murali. Tegge has been working on the research for nearly two years, and wants to use the fellowship to move into an academic research track, becoming an independent researcher.

Tegge obtained her bachelor’s and master’s degrees from the University of Illinois in 2006 and 2008, respectively, and her doctoral degree in informatics from University of Missouri, Columbia, in 2012.



Daphne Yao
Blacksburg, VA , June 03, 2014
Virginia Tech College of Engineering

Cybercrime comes in all forms these days. One recent headline told of the creepware or silent computer snooping that resulted in the arrest of some 90 people in 19 countries.  Miss Teen USA was among the victims. Her computer had been turned into a camera and used to spy on her in her own bedroom.

On the commercial front, Target suffered the largest retail hack in U.S. history during the Christmas shopping season of 2013, and now the Fortune 500 company’s outlook is bleak with steep drops in profits.

New research to be announced at the June 2014 ACM Symposium on Information, Computer and Communications Security in Kyoto, Japan has unveiled the causal relations among computer network events.

The work effectively isolates infected computer hosts and detects in advance stealthy malware also known as malicious software. 

The work was conducted under the auspices of a 2010 National Science Foundation CAREER Award grant to develop software that differentiates human-user computer interaction from malware. That $530,000 award was presented to Danfeng (Daphne) Yao,  associate professor of computer science at Virginia Tech. She worked with Naren Ramakrishnan, the Thomas L. Phillips Professor of Engineering, and her graduate student Hao Zhang of Beijing, China, a doctoral candidate in computer science.

The Virginia Tech computer scientists used causal relations to determine whether or not network activities have justifiable and legitimate causes to occur.

“This type of semantic reasoning is new and very powerful,” Yao said.

“The true significance of this security approach is its potential proactive defense capability. Conventional security systems scan for known attack patterns, which is reactive. Our anomaly detection based on enforcing benign properties in network traffic is a clear departure from that,” Yao added.

They will present their paper "Detection of Stealthy Malware Activities with Traffic Causality and Scalable Triggering Relation Discovery" on June 4.  It will be published in the symposium’s proceedings.

Virginia Tech Intellectual Property has filed a patent on this technology, and it is actually a continuation-in-part patent, following one of Yao’s earlier patents.

Previously, Yao garnered a 3-year, $450,000 grant from the Office of Naval Research (ONR) on cyber security to quantitatively detect anomalies in Department of Defense (DOD) computers, mobile devices, command and control servers, and embedded systems deployed on navy ships.

Yao’s career research focus has been on this methodology development for novel, practical, and quantitative anomaly detection. Specifically, she is analyzing causal relations of events and producing instructions for detecting anomalies in computer programs, systems, and networks.



Virginia Tech researchers have created a new augmented reality app that would bring inquiry learning about historic sites to school children. The team is testing the app at the Christiansburg (Va.) Institute, a school founded after the Civil War. App developer team members are, left to right, Aaron Johnson, David Hicks, Rosemary Zlokas, Gurjot Singh, David Cline, Todd Ogle, and Doug Bowman.
Blacksburg, VA , May 09, 2014
Virginia Tech College of Engineering

Virginia Tech researchers are testing a new mobile software app that will allow school children visiting historic sites to learn in new, interactive ways via augmented reality.

The app will use augmented reality -- where the user’s view of the real world is enhanced with additional objects and information on a viewing screen -- to give students the ability to “see” a historic structure as it appeared decades ago through the camera lens of a smartphone or tablet. Users can explore the site and examine various objects as would a historian, looking for points of historical interest.

Further, children using the app will be able to read historic documents, view photographs, and listen to audio such as oral history interviews related to the location they are visiting. Students then can later analyze the material using a set of guiding questions, according to the app development team which encompasses Virginia Tech’s College of Engineering and College of Liberal Arts and Humanities, in addition to the Technology-enhanced Learning and Online Strategies program.

“The goal is to use technology to help secondary school students learn how to use evidence to build an interpretation of the past,” said David Hicks, associate professor of history and social science education. He likens young users of the education app to taking on the role of “junior history detectives,” tracking historical clues and building a story of what happened in the past.

How does the app work? It takes historic photographs, film, or other media and overlays them on the view screen of the smart mobile device as the user holds it up to an historic building or to a now empty  landscape where a structure once stood. The phone’s built-in camera, GPS, and other sensors allow the app to identify what or where the user is looking at, so that users can see how a historic building and its surroundings looked decades ago, or they can see razed buildings digitally appear where they once existed.

“It’s a new way to study history, to become a historian,” said Doug Bowman, a professor of computer science. Bowman is leading the augmented reality and user interface aspects of the project.

To prototype the app, the team chose the local Christiansburg Institute site, a historic African-American school in Christiansburg, Va., which prospered for more than a century following the Civil War. Now, only a few of the buildings on what had been a 180-acre campus remain. Culling old photographs, the app gives users the chance to “explore” buildings no longer standing while also providing access to previously inaccessible historical sources on site, or see inside the now boarded-up Edgar A. Long building.

User trials at the site – with local fifth-grade schoolteachers from the Montgomery County Public School system, and Virginia Tech graduate students – are planned for fall 2014. Development of the app is funded by the National Science Foundation for $549,000 during a two-year period.

“This app is meant to guide the student through a historic site and the process of asking questions and learning in much the same way that a teacher or docent would, but with the capacity to allow the user to instantly link to multiple historical sources, serve them up, and save them for later analysis,” said David Cline, an assistant professor of history.

The idea for a virtual history app as a learning tool came from the combined efforts of Hicks, Todd Ogle, a senior director with the Technology-enhanced Learning program, and Eric Ragan, now a researcher at Tennessee’s Oak Ridge National Laboratory, while he was a computer science doctoral student at Virginia Tech. At the time Ragan was writing his dissertation on educational virtual environments and obtaining information through on-site learning. Each seeking an interest in using augmented reality to view historical content in the context of the real world, the trio soon brought in Bowman and Cline.

“The augmented reality application is exciting, but the project is about more than just making a slick educational tool,” said Ragan. “It’s about studying new and effective ways to teach people to think critically and evaluate evidence.”

Further, leaving a historic site does not mean the end of learning, said Bowman, as the team is devising curriculum material to be used before and after site visits. “Students could use the mobile device to collect, annotate, and organize data at the local site, and then explore, manipulate, and analyze the data in the classroom.”

Designed to serve as a model for future apps, the platform could reshape historic tourism, leaving behind static signage or prerecorded digital audiotapes, making the experience of visiting landmarks an interactive, evolving experience for adults and students alike. The app also could find use in such fields as solving engineering problems, as well as the humanities and creative endeavors, such as art museums.

“I do see this being how we all interact with historic sites in the future; hopefully the near future,” said Ogle, adding that smaller research projects and commercial projects that use similar virtual apps for tourism are starting to emerge in various fields.

Additional members of the app team include Aaron Johnson of Sanford, N.C., a doctoral student with the School of Education’s Department of Teaching and Learning; Charles Layman, an augmented reality developer with Virginia Tech’s information technology office; Gurjot Singh, a post-doctoral researcher with the computer science department; Mohammedkhair “K.” Alnajar of Warrenton, Va., an undergraduate in computer science; and Rosemary Zlokas of Pittsburgh, a master’s student in history.



The team members, pictured from left to right, standing, are: Cristian Moral Martos, of Madrid, Spain; Mahdi Nabiyouni, of Tehran, Iran; and Doug Bowman, faculty adviser and professor of computer science at Virginia Tech. Seated is Felipe Bacim, of Porto Alegre, Brazil.
Blacksburg, VA , April 22, 2014
Virginia Tech College of Engineering

For the fourth time in five years, a team of Virginia Tech doctoral students from the College of Engineering’s Computer Science Department and Center for Human-Computer Interaction has won the top prize in the IEEE 3-D User Interfaces contest.

The contest, sponsored by the Computer Society of the Institute for Electrical and Electronics Engineers, was held at the 2014 Symposium on 3-D User Interfaces in Minneapolis, Minn. This year’s competition focused on three-dimensional point clouds, dense sets of points in 3-D space.

Point clouds are commonly produced by devices such as laser scanners. For example, a building can be scanned to obtain a detailed 3-D geometric model. The contest required teams to design and build systems for labeling such point clouds. This challenging task involves the precise selection of regions of points in 3-D space so that those points can be labeled.

The Virginia Tech team devised a solution entitled “Slice-n-Swipe” that provided multiple virtual tools to allow users to annotate point clouds using natural in-air gestures. For example, the “chef’s knife” tool allowed the user to slice through the point cloud in mid-air, and then swipe away the unwanted points. Further slicing and swiping can refine the selection until only the desired points remain.

Additional tools included a resizable “bubble” that could be used to paint the desired points and a “lasso” similar to the tool used in painting and photo editing software. The system was implemented using the Leap Motion Controller, a new input technology that can track the positions of the user’s hands and fingers in mid-air, and a 3-D mouse for controlling the user’s view of the point cloud. The team members produced a YouTube video describing the entire system.

“Our design was inspired by the possibilities for natural gesture-based interaction with devices like the Leap Motion Controller,” said Doug Bowman, faculty advisor to the team and professor of computer science. “The challenge was to design a precise and usable interface based on in-air motions that are inherently imprecise. We achieved this by designing a variety of tools and by using the concept of progressive refinement, where an initial rough selection is refined step-by-step until the precise result is accomplished.”

Computer science team members were: Felipe Bacim of Porto Alegre, Brazil; Mahdi Nabiyouni of Tehran, Iran; and Cristian Moral Martos, a visiting Ph.D. student from the Polytechnic University of Madrid in Spain. Bacim has been a part of all four of Virginia Tech’s winning teams in the 3-D UI contest (2010, 2011, 2012, and 2014).

At the symposium, the team competed against four other finalists from around the world, including two teams from Brazil, one team from Germany, and one from France. Judging was done based on live demonstrations over two days, and the winner was determined by a popular vote of the symposium attendees, including top 3-D user interface researchers from around the world.

In addition to the 3-D UI contest award, Nabiyouni, Bowman, and Bireswar Laha, a Ph.D. student in computer science from Konnagar, India, received an honorable mention best poster award for their poster, “Designing Effective Travel Techniques with Bare-Hand Interaction.” This research also used the Leap Motion Controller and investigated how to enable navigation through 3-D environments based on in-air gestures.


Syndicate content