Abstract: Techniques for classifying one or more objects in at least one video, wherein the at least one video comprises a plurality of frames are provided. One or more objects in the plurality of frames are tracked. A level of deformation is computed for each of the one or more tracked objects in accordance with at least one change in a plurality of histograms of oriented gradients for a corresponding tracked object. Each of the one or more tracked objects is classified in accordance with the computed level of deformation.

Abstract: A trajectory of movement of an object is tracked in a video data image field that is partitioned into a plurality of different grids. Global image features from video data relative to the trajectory are extracted and compared to a learned trajectory model to generate a global anomaly detection confidence decision value as a function of fitting to the learned trajectory model. Local image features are also extracted for each of the image field grids that include object trajectory, which are compared to learned feature models for the grids to generate local anomaly detection confidence decisions for each grid as a function of fitting to the learned feature models for the grids. The global anomaly detection confidence decision value and the local anomaly detection confidence decision values for the grids are into a fused anomaly decision with respect to the tracked object.

Abstract: The present invention involves implementation of an intelligent switching program, whereby the processing power required to monitor check-out stations is considerably reduced. The present invention monitors a subset of check-out stations at any given time, instead of monitoring all check-out stations at all times. The subset of check-out stations is determined dynamically according to, but not limited to, cashier records, input parameters from the user, current lane activity, past lane activity, time of day, etc. Statistical models (e.g., effective population sampling and/or population hypothesis tests) are developed along these lines that guide the lane selection process, whereby increases in the false-negative rate due to failure to monitor particular lanes when events of interest occur are controlled. By monitoring fewer check-out stations, while maintaining target performance accuracy, the amount of data that end users must deal with is significantly reduced.

Abstract: Techniques for detecting one or more events are provided. The techniques include using multiple overlapping regions of interest on a video sequence to cover a location for one or more events, wherein each event is associated with at least one of the multiple overlapping regions of interest, applying multiple-instance learning to the video sequence to select one or more of the multiple overlapping regions of interest to construct one or more location-aware event models, and applying the models to the video sequence to detect the one or more events and to determine the one or more regions of interest that are associated with the one or more events.

Abstract: Techniques for detecting one or more events are provided. The techniques include using one or more regions of interest on a video sequence to cover a location for one or more events, wherein each event is associated with at least one of the one or more regions of interest, applying multiple-instance learning to the video sequence to construct one or more location-aware event models, and applying the models to the video sequence to determine the one or more regions of interest that are associated with the one or more events.

Abstract: Techniques for classifying one or more objects in at least one video, wherein the at least one video comprises a plurality of frames are provided. One or more objects in the plurality of frames are tracked. A level of deformation is computed for each of the one or more tracked objects in accordance with at least one change in a plurality of histograms of oriented gradients for a corresponding tracked object. Each of the one or more tracked objects is classified in accordance with the computed level of deformation.

Abstract: Techniques for analyzing one or more sequential events performed by a human actor to evaluate efficiency of the human actor are provided. The techniques include identifying one or more segments in a video sequence as one or more components of one or more sequential events performed by a human actor, integrating the one or more components into one or more sequential events by incorporating a spatiotemporal model and one or more event detectors, and analyzing the one or more sequential events to analyze behavior of the human actor.

Abstract: An approach that detects objects crossing a virtual boundary line is provided. Specifically, an object detection tool provides this capability. The object detection tool comprises a boundary component configured to define a virtual boundary line in a video region of interest, and establish a set of ground patch regions surrounding the virtual boundary line. The object detection tool further comprises an extraction component configured to extract a set of attributes from each of the set of ground patch regions, and update a ground patch history model with the set of attributes from each of the set of ground patch regions. An analysis component is configured to analyze the ground patch history model to detect whether an object captured in at least one of the set of ground patch regions is crossing the virtual boundary line in the video region of interest.

Abstract: The present invention is directed to the measurement of attributes of a queue.

Abstract: Multi-mode video event indexing includes determining a quality of object distinctiveness with respect to images from a video stream input. A high-quality analytic mode is selected from multiple modes and applied to video input images via a hardware device to determine object activity within the video input images if the determined level of detected quality of object distinctiveness meets a threshold level of quality, else a low-quality analytic mode is selected and applied to the video input images via a hardware device to determine object activity within the video input images, wherein the low-quality analytic mode is different from the high-quality analytic mode.

Abstract: Human behavior is determined by sequential event detection by constructing a temporal-event graph with vertices representing adjacent first and second primitive images of a plurality of individual primitive images parsed from a video stream, and also of first and second idle states associated with the respective first and second primitive images. Constructing the graph is a function of an edge set between the adjacent first and second primitive images, and an edge weight set as a function of a discrepancy between computed visual features within regions of interest common to the adjacent first and second primitive images. A human activity event is determined as a function of a shortest distance path of the temporal-event graph vertices.

Abstract: The present invention involves implementation of an intelligent switching program, whereby the processing power required to monitor check-out stations is considerably reduced. The present invention monitors a subset of check-out stations at any given time, instead of monitoring all check-out stations at all times. The subset of check-out stations is determined dynamically according to, but not limited to, cashier records, input parameters from the user, current lane activity, past lane activity, time of day, etc. Statistical models (e.g., effective population sampling and/or population hypothesis tests) are developed along these lines that guide the lane selection process, whereby increases in the false-negative rate due to failure to monitor particular lanes when events of interest occur are controlled. By monitoring fewer check-out stations, while maintaining target performance accuracy, the amount of data that end users must deal with is significantly reduced.

Abstract: A system, method and program product for providing automated learning for a people counting system. A system is disclosed that includes a grid system for dividing a field of view (FOV) of a captured image data into a set of blocks; an object detection and tracking system for tracking a blob passing through the FOV; and a learning system that maintains person size parameters for each block and updates person size parameters for a selected block when a blob appears in the selected block.

Abstract: Techniques for using transactional and visual event information to facilitate loss prevention are provided. The techniques include obtaining video of one or more visual events at a point of sale environment and one or more transaction log entries that correspond to the video, wherein the one or more transaction log entries comprise one or more transactional events, categorizing each event as one of one or more model events, using each categorized event to create a revised transaction log, wherein the revised transaction log comprises a sequence of categorized events, wherein each categorized event is a combination of the one or more transactional events and the one or more visual events, examining the revised transaction log to correct one or more mis-categorizations, if any, and revise one or more model event categories with the one or more corrected mis-categorizations, if any, and using the revised transaction log to facilitate loss prevention.

Abstract: Techniques for analyzing one or more sequential events performed by a human actor to evaluate efficiency of the human actor are provided. The techniques include identifying one or more segments in a video sequence as one or more components of one or more sequential events performed by a human actor, integrating the one or more components into one or more sequential events by incorporating a spatiotemporal model and one or more event detectors, and analyzing the one or more sequential events to analyze behavior of the human actor.

Abstract: Techniques for detecting one or more events are provided. The techniques include identifying one or more segments in a video sequence as one or more candidates for one or more events by a temporal ordering of the one or more candidates, and analyzing one or more motion patterns of the one or more candidates to detect the one or more events.

Abstract: Techniques for detecting one or more events are provided. The techniques include using one or more regions of interest on a video sequence to cover a location for one or more events, wherein each event is associated with at least one of the one or more regions of interest, applying multiple-instance learning to the video sequence to construct one or more location-aware event models, and applying the models to the video sequence to determine the one or more regions of interest that are associated with the one or more events.

Abstract: An approach that allows for model based people counting is provided. In one embodiment, there is a generating tool configured to generate a set of person-shape models based on results of a cumulative training process; a detecting tool configured to detect persons in a camera field-of-view by using the set of person-shape models, and a counting tool configured to track detected persons upon crossing by the detected persons of a previously established virtual boundary.

Abstract: An approach that detects objects crossing a virtual boundary line is provided. Specifically, an object detection tool provides this capability. The object detection tool comprises a boundary component configured to define a virtual boundary line in a video region of interest, and establish a set of ground patch regions surrounding the virtual boundary line. The object detection tool further comprises an extraction component configured to extract a set of attributes from each of the set of ground patch regions, and update a ground patch history model with the set of attributes from each of the set of ground patch regions. An analysis component is configured to analyze the ground patch history model to detect whether an object captured in at least one of the set of ground patch regions is crossing the virtual boundary line in the video region of interest.

Abstract: An approach that detects potentially fraudulent transactions is provided. In one embodiment, there is a fraud detection tool including, an identification component configured to identify a first person present within a zone of interest at a point of sale (POS) device using a set of sensor devices; a transaction component configured to determine whether the POS device has performed a first transaction and a second transaction while the first person is present within the zone of interest at the POS device; an analysis component configured to: analyze a transaction type of the first transaction and the second transaction; and detect whether the second transaction is potentially fraudulent based on a determination of whether the POS device has performed a first transaction and a second transaction while the first person is within the zone of interest at the POS device, and an analysis of the transaction type of the second transaction.