Computer_Vision_and_Image_Understanding.pdfVIP

Computer Vision and Image Understanding 113 (2009) 291–307 Contents lists available at ScienceDirect Computer Vision and Image Understanding journal homepage: www.else /l ocate/cviu Detecting motion patterns via direction maps with application to surveillance Jacob M. Gryn *, Richard P. Wildes, John K. Tsotsos Department of Computer Science Engineering, Centre for Vision Research, York University, 4700 Keele St., Toronto, Ont., Canada M3J 1P3 a r t i c l e i n f o a b s t r a c t Article history: Detection of motion patterns in video data can be significantly simplified by abstracting away from pixel Received 13 July 2005 intensity values towards representations that explicitly and compactly capture movement across space Accepted 23 October 2008 and time. A novel representation that captures the spatiotemporal distributions of motion across regions Available online 13 November 2008 of interest, called the ‘‘Direction Map,” abstracts video data by assigning a two-dimensional vector, rep- resentative of local direction of motion, to quantized regions in space-time. Methods are presented for Keywords: recovering direction maps from video, constructing direction map templates (defining target motion pat- Surveillance terns of interest) and comparing templates to newly acquired video (for pattern detection and localiza- Direction maps Dominant direction tion). These methods