MONDAY 21.11.2011, 3.30pm at Bilik Mesyuarat 1, FTSM.
Abstract:
The talk will describe work that was done as part of a program of research funded by Transport for London to detect and classify the various road users (vehicles, cyclists, pedestrians, etc.) in London
streets. There are more than 1500 cameras covering key roads in the city, which are very difficult to monitor manually. Automatic processing is highly desirable for traffic enforcement, management and
planning. A detector and classifier for road users from images is presented that combines the concept of 3D models with Histogram of Oriented Gradients (HOG) to overcome limitations of conventional
silhouette-based methods for road user classification. The appearance of road users varies substantially with the viewing angle and would require several single view object detectors like HOG. Here, 3D models are used for the object categories to be detected to train a single and complete 3D appearance model applicable for any viewing angle. A calibrated camera allows an affine transformation of the observation (video frame) into a normalised representation; features (3D extended
histogram of oriented gradients) are defined. A variable set of interest points is used in the detection and classification processes, depending on which points in the 3D model are visible.The 3DHOG
feature is compared with features based on FFT and basic histograms and also a motion silhouette baseline on the reference benchmark i-LIDS data set from the UK Home Office. "
Biography:
Sergio A. Velastin received the B.Sc. degree in Electronics, M.Sc. (Research) degree in digital Image Processing and the Ph.D. from the University of Manchester Institute of Science and Technology (UMIST) in the UK, in 1978, 1979 and 1982 respectively. Currently he is a Professor of Applied Computer Vision and Director of the Digital Imaging Research Centre. His research interests include computer vision for pedestrian monitoring and personal security, transport monitoring and
embedded/distributed visual surveillance systems. He is an elected member of the Board of Governors of the IEEE's Society of Intelligent Transportation Systems. Prof Velastin is also a member of the IET and the British Machine Vision Association (BMVA).
