October 2008 Archives

I attended Dr. Andrew Zisserman's keynote talk on visual search, the VideoGoogle, in the Asian Conference on Computer Vision in 2006. With Google being an expert in text search, a similar application for videos is definitely the next-gen search application. The talk was magic, and novices like me were held captive when he presented a visual search demo on locating objects precisely in a movie, irrespective of the location, scale and viewing direction. It was finding a needle in the haystack. His infectious enthusiasm, willingness to discuss ideas with experts and novice students alike, and an extremely down-to-earth nature definitely won him many hearts at the conference. I, at least, was very inspired and drawn to the problem of visual search. Since then, I have been following upon Dr. Zisserman's work in detail and I find it very much fundamental, yet very applied to real-world problems.

Dr. Andrew Zisserman is currently a professor at the University of Oxford. To talk about his impact in a non-subjective manner is easy - perhaps 1 sentence long - "Andrew Zisserman has a Hirsch-index of 40". The Hirsch-index or the h-index as it is more commonly known, is a quantification of the actual scientific productivity and the apparent scientific impact of a scientist. A value of 40 means that 40 of Dr. Zisserman's publications have been cited by at least 40 other scientific publications each. Talk about impact! He has contributed to theoretical computer vision work as well as some very applied problems. He has co-authored a textbook "Multiple View Geometry in Computer Vision" with Dr. Richard Hartley. He continues to influence students and researchers across the world.

Long live Andrew Zisserman!

A number of people will argue that Ph.D. students, especially those in technical fields, lack a decent social life. Well, I am certainly not going to risk making any such controversial statements. On the other hand, I do believe that a large part of our social life stems from professional relationships that are formed at different conferences and gatherings of professional communities. I have made some great friends at conferences and I enjoy corresponding/ meeting with them every now and then.

I am already a student member of the ACM (Association for Computing Machinery) which is
the world's largest educational and scientific computing society. I am particularly involved in the activities of SIGMM which is a special interest group in Multimedia. This community strives to build cutting edge multimedia technology through publications, conferences and special workshops that provide a platform for researchers world-wide to interact.

Another leading association is the IEEE (Institute of Electrical and Electronic Engineers) which has similar structure of conferences, journals and workshops. What is different about the IEEE is that it also defines standards and measures for various technologies. The ACM and IEEE digital libraries are probably the highest visited sites by scientific community and are really a boon for students and researchers alike. By bringing global research to one's doorstep, they have truly enabled scientific progress like never before.

The impact of these organizations is even more visible in conferences. I attended the ACM Multimedia conference in 2006 and 2007 where I got to interact with leading researchers in multimedia community. ACM Multimedia is a conference that gives a flavor of all aspects of multimedia computing: from underlying technologies to applications, theory to practice, and servers to networks to devices. Other prominent conferences that I am interested in, are ICCV (International Conference on Computer Vision), ICPR (International Conference on Pattern Recognition), CVPR (Computer Vision and Pattern Recognition) and ACCV (Asian Conference on Computer Vision). I remember listening to Dr. Andrew Zisserman's keynote talk at the ACCV 2006 on VideoGoogle. This was the talk that inspired me to look into the problem of multimedia retrieval in the first place. Over last 3-4 years, I have closely followed the above conferences and become really interested in the aspects of pattern recognition and multimedia retrieval. I believe they have played an important role in making me want to become a researcher.

 

Well, this week's assignment is to blog about a senior IST graduate student from our respective labs. Unfortunately, Dr. Wang currently does not have a student who is in third or fourth year of Ph.D. So, I have decided to introduce to you Saurabh Kataria, a third year IST student co-advised by Dr. Giles and Dr. Prasenjit Mitra. Before coming to Penn State, Saurabh was working as a software developer in GlobalLogic, a startup based in India. He has a  Bachelors degree in Computer Science from Institute of Technology, Banaras Hindu University.

Saurabh is currently working on extraction of data points from two-dimensional plots in digital documents. He has published four papers around this topic, two of them as the first author. He, like me, is interested in understanding the digital media and extracting useful information that can help create more utility for a digital artifact. Best wishes to him for a great career!

 

Complementing my earlier blog post, this blog is about the professional activities and achievements of my adviser, Dr. James Wang. Dr. Wang received Bachelor's degree in Mathematics and Computer Science from University of Minnesota, an M. S. in Mathematics and another M. S. in Computer Science, both from Stanford University. He has a Ph. D. in Medical Information Sciences from Stanford University.  He is now an associate professor of the College of Information Sciences and Technology, the Computer Science and Engineering department, and the Integrative Biosciences (IBIOS) Program at the Pennsylvania State University. He is also the Vice Director of the Intelligent Information Systems Laboratory of which I am a member. He has been a Visiting Professor of the Robotics Institute, Carnegie Mellon University in 2007-2008. 
 
Dr. Wang specializes in visual database search and retrieval. He is the co-developer of two of the most famous image retrieval systems - the SIMPLIcity (semantics-sensitive image retrieval system) and the ALIPR (automatic linguistic indexing of pictures - real time).

Dr. Wang has authored two monographs, nearly 100 journal articles, book chapters, and many conference papers. But his emphasis has always been on publishing in top-tier journals. There is a Wikipedia entry on him. The best places to find more details about Dr. Wang are his website and the introduction in Penn State's directory.