Abstract: A method and system for detecting anomalies in video footage. A training dictionary can be configured to include a number of event classes, wherein events among the event classes can be defined with respect to n-dimensional feature vectors. One or more nonlinear kernel function can be defined, which transform the n-dimensional feature vectors into a higher dimensional feature space. One or more test events can then be received within an input video sequence of the video footage. Thereafter, a determination can be made if the test event(s) is anomalous by applying a sparse reconstruction with respect to the training dictionary in the higher dimensional feature space induced by the nonlinear kernel function.

Abstract: Methods, systems, and processor-readable media for video anomaly detection based upon a sparsity model. A video input can be received and two or more diverse descriptors of an event can be computed from the video input. The descriptors can be combined to form an event matrix. A sparse reconstruction of the event matrix can be performed with respect to an over complete dictionary of training events represented by the diverse descriptors. A step can then be performed to determine if the event is anomalous by computing an outlier rejection measure on the sparse reconstruction.

Abstract: Methods and systems for detecting anomalies in transportation related video footage. In an offline training phase, receiving video footage of a traffic location can be received. Also, in an offline training phase, event encodings can be extracted from the video footage and collected or compiled into a training dictionary. One or more input video sequences captured at the traffic location or a similar traffic location can be received in an online detection phase. Then, an event encoding corresponding to the input video sequence can be extracted. The event encoding can be reconstructed with a low rank sparsity prior model applied with respect to the training dictionary. The reconstruction error between actual and reconstructed event encodings can then be computed in order to determine if an event thereof is anomalous by comparing the reconstruction error with a threshold.

Abstract: Methods and systems for automatically detecting multi-object anomalies at a traffic intersection utilizing a joint sparse reconstruction model. A first input video sequence at a first traffic location can be received and at least one normal event involving P moving objects (where P is greater than or equal to 1) can be identified in an offline training phase. The normal event in the first input video sequence can be assigned to at least one normal event class and a training dictionary suitable for joint sparse reconstruction can be built in the offline training phase. A second input video sequence captured at a second traffic location similar to the first traffic location can be received and at least one event involving P moving objects can be identified in an online detection phase.

Abstract: Methods, systems, and processor-readable media for pruning a training dictionary for use in detecting anomalous events from surveillance video. Training samples can be received, which correspond to normal events. A dictionary can then be constructed, which includes two or more classes of normal events from the training samples. Sparse codes are then generated for selected training samples with respect to the dictionary derived from the two or more classes of normal events. The size of the dictionary can then be reduced by removing redundant dictionary columns from the dictionary via analysis of the sparse codes. The dictionary is then optimized to yield a low reconstruction error and a high-interclass discriminability.

Abstract: Methods, systems, and processor-readable media for video anomaly detection based upon a sparsity model. A video input can be received and two or more diverse descriptors of an event can be computed from the video input. The descriptors can be combined to form an event matrix. A sparse reconstruction of the event matrix can be performed with respect to an over complete dictionary of training events represented by the diverse descriptors. A step can then be performed to determine if the event is anomalous by computing an outlier rejection measure on the sparse reconstruction.

Abstract: Methods and systems for detecting anomalies in transportation related video footage. In an offline training phase, receiving video footage of a traffic location can be received. Also, in an offline training phase, event encodings can be extracted from the video footage and collected or compiled into a training dictionary. One or more input video sequences captured at the traffic location or a similar traffic location can be received in an online detection phase. Then, an event encoding corresponding to the input video sequence can be extracted. The event encoding can be reconstructed with a low rank sparsity prior model applied with respect to the training dictionary. The reconstruction error between actual and reconstructed event encodings can then be computed in order to determine if an event thereof is anomalous by comparing the reconstruction error with a threshold.

Abstract: Methods, systems, and processor-readable media for pruning a training dictionary for use in detecting anomalous events from surveillance video. Training samples can be received, which correspond to normal events. A dictionary can then be constructed, which includes two or more classes of normal events from the training samples. Sparse codes are then generated for selected training samples with respect to the dictionary derived from the two or more classes of normal events. The size of the dictionary can then be reduced by removing redundant dictionary columns from the dictionary via analysis of the sparse codes. The dictionary is then optimized to yield a low reconstruction error and a high-interclass discriminability.

Abstract: A method and system for detecting anomalies in video footage. A training dictionary can be configured to include a number of event classes, wherein events among the event classes can be defined with respect to n-diminensional feature vectors. One or more nonlinear kernel function can be defined, which transform the n-dimensional feature vectors into a higher dimensional feature space. One or more test events can then be received within an input video sequence of the video footage. Thereafter, a determination can be made if the test event(s) is anomalous by applying a sparse reconstruction with respect to the training dictionary in the higher dimensional feature space induced by the nonlinear kernel function.

Abstract: A method, system and processor-readable medium for automatically detecting anomalies at a traffic intersection. A set of clusters of nominal vehicle paths and a set of clusters of nominal trajectories within the nominal vehicle paths can be derived in an offline process. A set of features within each nominal trajectory among the set of clusters of nominal trajectories can be selected. A probability distribution for features indicative of nominal vehicle behavior within the nominal trajectories can be derived. An input video sequence can be received and presence of the anomaly in the vehicle path, trajectories and features within the input video sequence can be detected utilizing the derived path clusters, trajectory clusters, and feature distributions.

Abstract: Methods and systems for automatically detecting multi-object anomalies at a traffic intersection utilizing a joint sparse reconstruction model. A first input video sequence at a first traffic location can be received and at least one normal event involving P moving objects (where P is greater than or equal to 1) can be identified in an offline training phase. The normal event in the first input video sequence can be assigned to at least one normal event class and a training dictionary suitable for joint sparse reconstruction can be built in the offline training phase. A second input video sequence captured at a second traffic location similar to the first traffic location can be received and at least one event involving P moving objects can be identified in an online detection phase.