Abstract: Systems and methods for audience monitoring are provided that include receiving an input including a recording or live feed of an audience composed of several persons, detecting foreground of the input, performing blob segmentation of the input, and analyzing human presence on each segmented blob by identifying at least one person, identifying a spatial distribution of at least one identified person, determining a dwell time of at least one identified person, determining a temporal distribution of at least one identified person, and determining a gaze direction of at least one identified person. Such detecting provides the ability to track individual persons present in the audience, and how long they remain in the audience. The method also provides the ability to determine gaze direction of persons in the audience, and how long one or more persons are gazing in a particular direction.

Abstract: Methods and systems are provided to determine a target tracking box that surrounds a moving target. The pixels that define an image within the target tracking box can be classified as background pixels, foreground pixels, and changing pixels which may include pixels of an articulation, such as a portion of the target that moves relatively to the target tracking box. Identification of background image pixels improves the signal-to-noise ratio of the image, which is defined as the ratio between the number of pixels belonging to the foreground to the number of changing pixels, and which is used to track the moving target. Accordingly, tracking of small and multiple moving targets becomes possible because of the increased signal-to-noise ratio.

Abstract: Method and system for objects surveillance and real-time activity recognition is based on analysis of spatio-temporal images of individuals under surveillance where a spatio-temporal volume occupied by each individual is decomposed by crossing the same at specific heights to form 2-dimensional slices, each containing representation of trajectory of the motion of corresponding portions of the individual body. The symmetry of the trajectories (Gait DNA) is analyzed and classified to generate data indicative of a type of activity of the individual based on the symmetry or asymmetry of the Gait DNA in each 2-dimensional slice. An effective occlusion handling ability is implemented which permits to restore the occluded silhouette of an individual.

Abstract: Methods and systems are provided to determine a target tracking box that surrounds a moving target. The pixels that define an image within the target tracking box can be classified as background pixels, foreground pixels, and changing pixels which may include pixels of an articulation, such as a portion of the target that moves relatively to the target tracking box. Identification of background image pixels improves the signal-to-noise ratio of the image, which is defined as the ratio between the number of pixels belonging to the foreground to the number of changing pixels, and which is used to track the moving target. Accordingly, tracking of small and multiple moving targets becomes possible because of the increased signal-to-noise ratio.

Abstract: Systems and methods for audience monitoring are provided that include receiving an input including a recording or live feed of an audience composed of several persons, detecting foreground of the input, performing blob segmentation of the input, and analyzing human presence on each segmented blob by identifying at least one person, identifying a spatial distribution of at least one identified person, determining a dwell time of at least one identified person, determining a temporal distribution of at least one identified person, and determining a gaze direction of at least one identified person. Such detecting provides the ability to track individual persons present in the audience, and how long they remain in the audience. The method also provides the ability to determine gaze direction of persons in the audience, and how long one or more persons are gazing in a particular direction.

Abstract: Method and system for objects surveillance and real-time activity recognition is based on analysis of spatio-temporal images of individuals under surveillance where a spatio-temporal volume occupied by each individual is decomposed by crossing the same at specific heights to form 2-dimensional slices, each containing representation of trajectory of the motion of corresponding portions of the individual body. The symmetry of the trajectories (Gait DNA) is analyzed and classified to generate data indicative of a type of activity of the individual based on the symmetry or asymmetry of the Gait DNA in each 2-dimensional slice. An effective occlusion handling ability is implemented which permits to restore the occluded silhouette of an individual.

Abstract: A method determines driver glance information from an input video by performing motion analysis on the video; and performing image analysis on a frame of the video.