Damian M. Lyons, Ph.D.
Associate Professor, Director, Robotics & Computer Vision Laboratory, and Chair, Department of Computer & Information Sciences, Fordham University, New York, New York

Dr. Lyons has undergraduate degrees in Math (B.A., 1980) and Electrical Engineering (B.A.I., 1980) and a Master's Degree in Computer Science (M.Sc., 1981) from Trinity College, University of Dublin, Ireland. He earned his doctorate in Computer Science from the University of Massachusetts at Amherst (Ph.D., 1986). Dr. Lyons is the director of Fordham's Robotics and Computer Vision Laboratory and his research interests include target tracking, camera handoff, multisensory fusion and behavior recognition. His background includes more than 15 years as a researcher and research program manager at the U.S. division of Philips Corporate Research; he was Department head for the Video and Display Processing research department, responsible for technical leadership and funding for this diverse group, and project leader for Philips' research activities in Automated Video Surveillance. He has served on numerous conference program committees, has published over 80 technical papers in conferences, journals and books, and is inventor/co-inventor of 13 U.S. patents. Dr. Lyons is a member of ACM and IEEE.


Automated CCTV surveillance presents many potential security benefits, however it is a challenging problem to solve. Difficulties include reversing the 3-Dimensional to 2-Dimensional camera transformation, handling varying background and lighting conditions, faithfully tracking arbitrarily moving targets through background and mutual occlusions, and segmenting and interpreting target activities. So while CCTV is potentially a very rich source of information, it can require a great deal of additional context information to be assumed or embedded into a standalone automated surveillance application. If on the other hand CCTV is treated as an integrated part of an overall cyber security system, then additional information can be brought to bear to interpret video imagery. Additional sources of information include sound, depth information, RFID data, PIR data, access control data, etc. We have developed a general and flexible approach to information fusion for CCTV target tracking that includes fusion of multiple cues and hence provides context to interpret video imagery.