Degrees: PHD, 1992, Univ Pennsylvania; MS, 1982, Univ Sofia

Bio:

Dr. Kamberova earned a M.S. in Mathematics, concentration Topology, from Sofia State University “St. Kliment Ochridski” in Bulgaria and a PhD in Computer Science from The University of Pennsylvania in Philadelphia. She is an alumni of the General Robotics and Active Sensory Perception Laboratory (GRASP Lab) at PENN.

Dr. Kamberova’s professional interests include computer vision, biometrics, multi-sensor fusion, and decision making under uncertainty. She has written and presented numerous papers on these topics at conferences sponsored by ACM, IEEE and SIAM. She has authored and co-authored several journal articles and book chapters on these topics, and served as an editor of a book on microarray image analysis published by DNA Press.

Dr. Kamberova’s most recent scholarly accomplishments are in the development of methods for reconstructing 3D scenes and shapes of objects from photographs , and in measuring the geometry of 3D point sets. This research has important applications, which include biometrics, medical imaging, homeland security, tele-immersion and virtual training and collaboration.

Dr. Kamberova came to Hofstra University as an assistant professor in 1999. She has been and associate professor and the chairperson of the Department since 2004. Before coming to Hofstra, Dr. Kamberova taught at The University of Pennsylvania and Washington University in St. Louis. She was a postdoctoral researcher at The University of Pennsylvania and Rice University. She has international research collaborations with Center for Machine Perception at The Czech Technical University.

Dr. Kamberova teaches variety of courses at undergraduate and graduate levels which include computer graphics, artificial intelligence and computer vision. Together with key faculty in the computer science department she has developed the Distance Learning MS program in computer science. Dr. Kamberova enjoys working with talented high school student, introducing them to research in computer vision. She has had several semifinalists on the Intel Talent Search Competition.

Dr. Kamberova had served as a program committee member of international conferences in computer vision organized by IEEE, ACM and Eurographics. Those include the International conference of computer vision and pattern recognition (CVPR), International Symposium of Visual Computing (ISVC), International Conference on Computer Vision Applications (VISAPP), and the international workshops on Applications computer vision ( WACAV) and the CVPR workshop “Beyond muliti-view geometry”. She has served as a reviewer for the IEEE Transcations on Pattern analysis and Machine Intelligence (PAMI), Journal of the Optical Society of America (JOSA A) , International Journal of Robotics, and the ACM Electronic Journal of Educational Resources in Computer Science (JERIC).

ACM student chapter sponsor

Research

3D topology and shape recovery of objects (scenes) from unorganized clouds of 3D points. This allows partitioning of the scene into different components (clustering points that are close to each other on the surface) and describing the "curvedness" at the points. This is very important for the purpose of efficient transmission of huge sets of 3D data, it also facilitates matching and registration of 3D target objects to a database of 3D objects.

Computer Vision. The subject of computer vision is the development of sensors and computational procedures to equip computers with the ability to "see". This might include: recovering the shape of the three dimensional objects in a scene from two dimensional photographs or from a video sequence; gathering of biometric measurements, for example from fingerprints and iris-prints, face measurements, or human gait measurements; detecting and classifying abnormal or dangerous behavior at public places or security areas; detecting tumors in medical scans images. Many sensors (laser range scanners, stereo cameras, CT scans) produce sets of 3D points that are sampled from the surfaces of the objects of scenes being observed. I am interested in recovering the shape of the objects from those unorganized clouds of points.

Multi-Sensor Fusion: the objective is to use multiple sensors and to integrate (fuse) the information so that the integrated data provide more complete and more reliable picture of the scene or process under investigation. Some examples include, generating a common 3D models form unorganized set of pictures taken with different cameras; tracking an object with multiple cameras as it moves across a scene where the individual cameras have only partial views of the scene; providing security for urban or military areas.

PhD Thesis: Robust minimax estimation for a class of highly non-Gausian distributions

Publications

Books:
DNA Array Image Analysis: Nuts and Bolts, G. Kamberova and S.Shah (eds), DNA Press, 2002.

Papers:
3D Geometry from Uncalibrated Images, G. Kamberov, G. Kamberova, O. Chum, S. Obdrzalek, D. Martinec, J. Kostkova, T. Pajdla, J. Matas, and R. Sara, 2nd Intl. Symposium on Visual Computing, Lake Tahoe, NV, Lecture Notes in Computer Science, 2006.

Recovering Surfaces from the Restoring Force, G. Kamberov, G. Kamberova, Proceedings of ECCV 2002. Lecture Notes in Comp. Science, Volume 2351, pp 598-612, Springer-Verlag Berlin Heidelberg 2002.

Minimax rules under zero-one loss for a restricted location parameter, G. Kamberova and M. Mintz, Journal of Statistical Planning and Inference, Volume 7, Issue 2, pp 205-221, 1999.

"Robust fusion of location data, G. Kamberova, R. Mandelbaum, M. Mintz and R. Bajcsy, invited paper, Journal of the Franklin Institute, special issue on sensor fusion and integration, Volume 336, No 2, pp 269 - 285, 1999.

Students: advising student research, including high-school student research which has yield Intel semifinalist.

Courses Taught:

Problem solving and programming, Data structures and Algorithms, Analysis of Algorithms Computer Graphics, Computer Vision, Machine Perception, Artificial Intelligence, Probability and Statistics, Design and Analysis of Experiments.

Independent studies with undergrads

• Stereo reconstruction.
• Recovery of 3D normals from unorganized point clouds.
• Multi-sensor fusion: video, acoustic and seismic sensors.
• Computer vision for the blind.

Honors thesis for undergrads

• Web based computer games to enhance the first grade curriculum.

Grad work supervised

• Enterprize File Storage Systems and Protocols.
• Image Data Compression and JPEG and MPEG file formats.
• Polynocular stereo camera system.
• Digital Forensics.