Recent Faculty News and Publications
Did you know that the computer science faculty are as passionate about their research as they are about teaching? Here is a list of the most recent faculty news and publications.
Andrea Salgian (Computer Science), M. Pfirrmann, T. M. Nakra
Research Paper: Follow the Beat? Understanding Conducting Gestures from Video
Presented at: 3rd International Symposium on Visual Computing, Lecture Notes in Computer Science Series, Springer, 2007, to appear.
Abstract: In this paper we present a vision system that analyzes the gestures of a noted conductor conducting a real orchestra, a different approach from previous work that allowed users to conduct virtual orchestras with prerecorded scores. We use a low-resolution video sequence of a live performance of the Boston Symphony Orchestra, and we track the conductor's right hand. The tracker output is lined up with the output of an audio beat tracker run on the same sequence. The resulting analysis has numerous implications for the understanding of musical expression and gesture.
Miroslav Martinovic (Computer Science), Srdan Vesic, Faculty of Mathematics, University of Belgrade (Serbia), Goran Rakic, Faculty of Mathematics, University of Belgrade (Serbia)
Research Paper: Building an Information Retrieval System for Serbian - Challenges and
Solutions
Presented at: INTERSPEECH 2007, Antwerp, Belgium, August 27-31
Abstract: We describe challenges encountered while building an information retrieval system for Serbian language. Approaches designed and adopted to handle them are depicted and illuminated in this paper. As a backbone of our system, we used SMART retrieval system which we augmented with features necessary to deal with specificities of the Serbian alphabet. In addition, morphological richness of the language accentuated implications of the text preprocessing phase. During this phase, we devised two algorithms which increased retrieval precision by 14% and 27%, respectively. Testing was conducted using two gigabyte EBART collection of Serbian newspaper articles.
Sarah Monisha Pulimood (Computer Science), Karen DelDuca and Alexandra Raymond (Computer Science Students)
Research Award: Designing the Next-Generation Magazine: Content, Usability and Information Security
Funding agency: Collaborative Research Experience for Undergraduates (CREU) in Computer Science and Engineering program
Andrea Salgian (Computer Science)
Research Paper: Using Multiple Patches for 3D Object Recognition
Presented at: The 2nd Beyond Patches Workshop, held in conjunction with the IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, MN, June 23, 2007
Abstract: Image patches have become increasingly popular in a variety of applications, due to their resistance to clutter and partial occlusion, as well as their partial insensitivity to object pose. Recently Mikolajczyk and Schmid compared a number of local descriptors and concluded that the SIFT-based ones perform best in image matching tasks. In this paper we analyze the performance of three patch descriptors in the context of 3D object recognition: SIFT , PCA-SIFT and keyed context patches. We use a data set containing images of six objects on clean and cluttered backgrounds, taken around the whole viewing sphere, and we look at individual and fused performances. Individually, the keyed context patches perform best overall, but they are outperformed for some objects by SIFT and PCA-SIFT. Recognition is improved by fusing the rankings generated by these classifiers.
Andrea Salgian (Computer Science), Andrew Zdybak, and Jaclyn English (Computer Science Students)
Research Paper: Towards an Automated End-To-End Face Recognition System
Presented at: The 2007 International Conference on Image Processing, Computer Vision, and Pattern Recognition (IPCV'07), Las Vegas, Nevada, June 25-28, 2007
Abstract: In addition to a face recognition algorithm, end-to-end face recognition systems require a feature detection stage, which usually includes face and eye detection. Each of these steps is a challenge in itself, and has been studied extensively by researchers. However, most of the work concentrated on detection or recognition alone (using manually detected features), and there has been little work on combining these stages into one system.
In this paper we describe and analyze an end-to-end face recognition system that includes automatic face detection, eye detection and face recognition. We look at the effect that localization error has on recognition performance. We conclude that while the decrease in performance is evident, thresholds can be set such that the human operator is alerted when the confidence of the system is low.
Sarah Monisha Pulimood (Computer Science) and Thomas R. Hagedorn (Mathematics)
Research Paper: Applying the Grid Computing Paradigm within a Liberal Arts Academic Environment
Presented at: The 2007 International Conference on Grid Computing and Applications (GCA'07), June 25-28, 2007
Abstract: Until very recently analysis and computations on a large scale were feasible only on supercomputers or clusters of high-end processors. Such computational infrastructure requires massive investments that can be unrealistic in a liberal arts college environment such as that found at The College of New Jersey (TCNJ). However, TCNJ has several state-of-the-art campus computer labs for use by students and faculty. There are periods ( e.g. nights, weekends, and summer months), when a significant number of these computers are underutilized. A grid computing environment would enable a user to harness underutilized campus resources for computationally intensive applications. In this paper we discuss some of the issues faced in the liberal arts environment, our approach for resolving them, and our initial results.
