Abstract: There is provided a computerized method for determining attention of an audience individual. The method comprises receiving a tracked sequence of an individual's attention directions, where the tracked sequence is indicative of a direction of attention of the one audience individual in each frame of a first subset of a first series of frames. The first series of frames are associated with a time interval. The method also comprises receiving a motion trajectory of a moving object, which is indicative of a location of a moving object in each frame of a second subset of a second series of frames. The second series of frames are associated with a second time interval that corresponds to the first time interval. The method further comprises processing the tracked sequence and the motion trajectory for determining an attention score of the audience individual towards the moving object.

Abstract: A method, system, and computer program product for stabilizing frames, the method comprising: receiving a frame sequence comprising three or more frames, including a current frame; determining salient feature points within the frames; matching the salient feature points between the frames; dropping salient feature points associated with advancing objects; dropping salient feature points associated with objects moving in shaking movements; computing a transformation between pairs of consecutive frames from amongst the at least three frames, based upon non-dropped salient feature points, thereby obtaining a multiplicity of transformations; determining a center position for the frames based upon the multiplicity of transformations; determining a stabilizing transformation from a current frame to the center position; and applying the stabilizing transformation to the current frame to obtain a stabilized frame.

Abstract: A method, system, and computer program product for stabilizing frames, the method comprising: receiving a frame sequence comprising three or more frames, including a current frame; determining salient feature points within the frames; matching the salient feature points between the frames; dropping salient feature points associated with advancing objects; dropping salient feature points associated with objects moving in shaking movements; computing a transformation between pairs of consecutive frames from amongst the at least three frames, based upon non-dropped salient feature points, thereby obtaining a multiplicity of transformations; determining a center position for the frames based upon the multiplicity of transformations; determining a stabilizing transformation from a current frame to the center position; and applying the stabilizing transformation to the current frame to obtain a stabilized frame.

Abstract: Methods and systems for determining the likelihood of a traffic collision between vehicles of different vehicle classes at an intersection are disclosed. Data informative of an intersection model is obtained. A first and second vehicle appearing in image data are classified, their trajectories extracted, and a plurality of references trajectories associated with their respective classes are selected from the intersection model. Conflicting pairs of references trajectories are identified. For each pair, the first vehicle is mapped to a point on the first reference trajectory and the second vehicle is mapped to a point on the second reference trajectory. Data indicative of the likelihood of a collision is generated, and a warning is generated when the generated data satisfy a predetermined criterion.

Abstract: A method, system, and computer program product for stabilizing frames, the method comprising: receiving a frame sequence comprising three or more frames, including a current frame; determining salient feature points within the frames; matching the salient feature points between the frames; dropping salient feature points associated with advancing objects; dropping salient feature points associated with objects moving in shaking movements; computing a transformation between pairs of consecutive frames from amongst the at least three frames, based upon non-dropped salient feature points, thereby obtaining a multiplicity of transformations; determining a center position for the frames based upon the multiplicity of transformations; determining a stabilizing transformation from a current frame to the center position; and applying the stabilizing transformation to the current frame to obtain a stabilized frame.