Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · 2008 · 131 citations · 14 references
ComputingAutonomous SystemsComputer ScienceComputer VisionMachine VisionTime Series AnalysisDigital SurveillanceEvent DetectionMotion AnalysisMotion EstimationOptical FlowVisual TrackingAspectAutomated Event DetectionMotion Feature DetectionMotion Field EstimationOptical Flow EstimationPoint TrackingSpatiotemporal Feature ExtractionVideo Anomaly Detection
Video-surveillance systems are becoming more and more autonomous in the detection and the reporting of abnormal events. In this context, this paper presents an approach to detect abnormal situations in crowded scenes by analyzing the motion aspect instead of tracking subjects one by one. The proposed approach estimates sudden changes and abnormal motion variations of a set of points of interest (POI). The number of tracked POIs is reduced using a mask that corresponds to hot areas of the built motion heat map. The approach detects events where local motion variation is important compared to previous events. Optical flow techniques are used to extract information such as density, direction and velocity. To demonstrate the interest of the approach, we present the results on the detection of collapsing events in real videos of airport escalator exits.
14
A Combined Corner and Edge Detector
Chris Harris, Matthew J. Stephens · 1988 · 12.5K citations
Jianbo Shi, Tomasi · 1994 · 6.9K citations · Full text
Foreground object detection from videos containing complex background
Liyuan Li, Weimin Huang, Irene Yu‐Hua Gu et al. · 2003 · 391 citations