Abstract: Techniques are disclosed for generating logical sensors for an image driver. The image driver monitors values corresponding to at least a first feature in one or more regions of a first image in a stream of images received by a first sensor. The image driver identifies at least a first correlation between at least a first and second value of the monitored values. The image driver generates a logical sensor based on the identified correlations. The logical sensor samples one or more features corresponding to the identified correlation from a second image in the stream of images.

Abstract: Techniques are disclosed for analyzing a scene depicted in an input stream of video frames captured by a video camera. Bounding boxes are determined for a set foreground patches identified in a video frame. For each bounding box, the techniques include determining textures for first areas, each including a foreground pixel and surrounding pixels, and determining textures for second areas including pixels of the background model image corresponding to the pixels of the foreground areas. Further, for each foreground pixel in the bounding box area, a correlation score is determined based on the texture of the corresponding first area and second area. Pixels whose correlation scores exceed a threshold are removed from the foreground patch. The size of the bounding box may also be reduced to fit the modified foreground patch.

Abstract: Techniques are disclosed for learning and modeling a background for a complex and/or dynamic scene over a period of observations without supervision. A background/foreground component of a computer vision engine may be configured to model a scene using an array of ART networks. The ART networks learn the regularity and periodicity of the scene by observing the scene over a period of time. Thus, the ART networks allow the computer vision engine to model complex and dynamic scene backgrounds in video.

Abstract: A machine-learning engine is disclosed that is configured to recognize and learn behaviors, as well as to identify and distinguish between normal and abnormal behavior within a scene, by analyzing movements and/or activities (or absence of such) over time. The machine-learning engine may be configured to evaluate a sequence of primitive events and associated kinematic data generated for an object depicted in a sequence of video frames and a related vector representation. The vector representation is generated from a primitive event symbol stream and a phase space symbol stream, and the streams describe actions of the objects depicted in the sequence of video frames.

Abstract: Embodiments presented herein describe a method for processing streams of data of one or more networked computer systems. According to one embodiment of the present disclosure, an ordered stream of normalized vectors corresponding to information security data obtained from one or more sensors monitoring a computer network is received. A neuro-linguistic model of the information security data is generated by clustering the ordered stream of vectors and assigning a letter to each cluster, outputting an ordered sequence of letters based on a mapping of the ordered stream of normalized vectors to the clusters, building a dictionary of words from of the ordered output of letters, outputting an ordered stream of words based on the ordered output of letters, and generating a plurality of phrases based on the ordered output of words.

Abstract: Systems and methods for viewing a scene depicted in a sequence of video frames and identifying and tracking objects between separate frames of the sequence. Each tracked object is classified based on known categories and a stream of context events associated with the object is generated. A sequence of primitive events based on the stream of context events is generated and stored together, along with detailed data and generalized data related to an event. All of the data is then evaluated to learn patterns of behavior that occur within the scene.

Abstract: A machine-learning engine is disclosed that is configured to recognize and learn behaviors, as well as to identify and distinguish between normal and abnormal behavior within a scene, by analyzing movements and/or activities (or absence of such) over time. The machine-learning engine may be configured to evaluate a sequence of primitive events and associated kinematic data generated for an object depicted in a sequence of video frames and a related vector representation. The vector representation is generated from a primitive event symbol stream and a phase space symbol stream, and the streams describe actions of the objects depicted in the sequence of video frames.

Abstract: Techniques are disclosed for creating a background model of a scene using both a pixel based approach and a context based approach. The combined approach provides an effective technique for segmenting scene foreground from background in frames of a video stream. Further, this approach can scale to process large numbers of camera feeds simultaneously, e.g., using parallel processing architectures, while still generating an accurate background model. Further, using both a pixel based approach and context based approach ensures that the video analytics system can effectively and efficiently respond to changes in a scene, without overly increasing computational complexity. In addition, techniques are disclosed for updating the background model, from frame-to-frame, by absorbing foreground pixels into the background model via an absorption window, and dynamically updating background/foreground thresholds.

Abstract: Techniques are disclosed for creating a background model of a scene using both a pixel based approach and a context based approach. The combined approach provides an effective technique for segmenting scene foreground from background in frames of a video stream. Further, this approach can scale to process large numbers of camera feeds simultaneously, e.g., using parallel processing architectures, while still generating an accurate background model. Further, using both a pixel based approach and context based approach ensures that the video analytics system can effectively and efficiently respond to changes in a scene, without overly increasing computational complexity. In addition, techniques are disclosed for updating the background model, from frame-to-frame, by absorbing foreground pixels into the background model via an absorption window, and dynamically updating background/foreground thresholds.

Abstract: A behavioral recognition system may include both a computer vision engine and a machine learning engine configured to observe and learn patterns of behavior in video data. Certain embodiments may provide image stabilization of a video stream obtained from a camera. An image stabilization module in the behavioral recognition system obtains a reference image from the video stream. The image stabilization module identifies alignment regions within the reference image based on the regions of the image that are dense with features. Upon determining that the tracked features of a current image is out of alignment with the reference image, the image stabilization module uses the most feature dense alignment region to estimate an affine transformation matrix to apply to the entire current image to warp the image into proper alignment.

Abstract: Techniques are disclosed for analyzing a scene depicted in an input stream of video frames captured by a video camera. The techniques include receiving data for an object within the scene and determining whether the object has remained substantially stationary within the scene for at least a threshold period. If the object is determined to have remained stationary for at least the threshold period, a rareness score is calculated for the object to indicate a likelihood of the object being stationary to an observed degree at an observed location. The rareness score may use a learning model to take into account previous stationary and/or non-stationary behavior of objects within the scene. In general, the learning model may be updated based on observed stationary and/or non-stationary behaviors of the objects. If the rareness score meets reporting conditions, the stationary object event may be reported.

Abstract: Techniques are disclosed for analyzing a scene depicted in an input stream of video frames captured by a video camera. The techniques include evaluating sequence pairs representing segments of object trajectories. Assuming the objects interact, each of the sequences of the sequence pair may be mapped to a sequence cluster of an adaptive resonance theory (ART) network. A rareness value for the pair of sequence clusters may be determined based on learned joint probabilities of sequence cluster pairs. A statistical anomaly model, which may be specific to an interaction type or general to a plurality of interaction types, is used to determine an anomaly temperature, and alerts are issued based at least on the anomaly temperature. In addition, the ART network and the statistical anomaly model are updated based on the current interaction.

Abstract: Embodiments presented herein describe techniques for generating a linguistic model of input data obtained from a data source (e.g., a video camera). According to one embodiment of the present disclosure, a sequence of symbols is generated based on an ordered stream of normalized vectors generated from the input data. A dictionary of words is generated from combinations of the ordered sequence of symbols based on a frequency at which combinations of symbols appear in the ordered sequence of symbols. A plurality of phrases is generated based an ordered sequence of words from the dictionary observed in the ordered sequence of symbols based on a frequency by which combinations of words in ordered sequence of words appear relative to one another.

Abstract: Techniques are disclosed for extracting micro-features at a pixel-level based on characteristics of one or more images. Importantly, the extraction is unsupervised, i.e., performed independent of any training data that defines particular objects, allowing a behavior-recognition system to forgo a training phase and for object classification to proceed without being constrained by specific object definitions. A micro-feature extractor that does not require training data is adaptive and self-trains while performing the extraction. The extracted micro-features are represented as a micro-feature vector that may be input to a micro-classifier which groups objects into object type clusters based on the micro-feature vectors.

Abstract: Techniques are disclosed for analyzing a scene depicted in an input stream of video frames captured by a video camera. The techniques include evaluating sequence pairs representing segments of object trajectories. Assuming the objects interact, each of the sequences of the sequence pair may be mapped to a sequence cluster of an adaptive resonance theory (ART) network. A rareness value for the pair of sequence clusters may be determined based on learned joint probabilities of sequence cluster pairs. A statistical anomaly model, which may be specific to an interaction type or general to a plurality of interaction types, is used to determine an anomaly temperature, and alerts are issued based at least on the anomaly temperature. In addition, the ART network and the statistical anomaly model are updated based on the current interaction.

Abstract: Techniques are disclosed for matching a current background scene of an image received by a surveillance system with a gallery of scene presets that each represent a previously captured background scene. A quadtree decomposition analysis is used to improve the robustness of the matching operation when the scene lighting changes (including portions containing over-saturation/under-saturation) or a portion of the content changes. The current background scene is processed to generate a quadtree decomposition including a plurality of window portions. Each of the window portions is processed to generate a plurality of phase spectra. The phase spectra are then projected onto a corresponding plurality of scene preset image matrices of one or more scene preset. When a match between the current background scene and one of the scene presets is identified, the matched scene preset is updated. Otherwise a new scene preset is created based on the current background scene.

Abstract: Techniques are disclosed for generating anomaly scores for a neuro-linguistic model of input data obtained from one or more sources. According to one embodiment, generating anomaly scores includes receiving a stream of symbols generated from an ordered stream of normalized vectors generated from input data received from one or more sensor devices during a first time period. Upon receiving the stream of symbols, generating a set of words based on an occurrence of groups of symbols from the stream of symbols, determining a number of previous occurrences of a first word of the set of words, determining a number of previous occurrences of words of a same length as the first word, and determining a first anomaly score based on the number of previous occurrences of the first word and the number of previous occurrences of words of the same length as the first word.

Abstract: Techniques are disclosed for analyzing and learning behaviors based on acquired sensor data. A neuro-linguistic cognitive engine performs learning and analysis on linguistic content (e.g., identified alpha symbols, betas, and gammas) obtained by a linguistic model that clusters observations to generate the linguistic content. The neuro-linguistic cognitive engine compares new data to learned patterns stored in short and longer-term memories and determines whether to issue special event notifications indicating anomalous behavior. In one embodiment, condition(s) may be generated for new data and checked against inference nodes of an inference network. Inference nodes matching the condition(s) are executed to, e.g., compare the new data with the learned patterns, with output from the inference nodes being used to generate additional condition(s) that are again matched to inference nodes which may be executed.

Abstract: Techniques are disclosed for analyzing and learning behavior in an acquired stream of video frames. In one embodiment, a trajectory analyzer clusters trajectories of objects depicted in video frames and builds a trajectory model including the trajectory clusters, a prior probability of assigning a trajectory to each cluster, and an intra-cluster probability distribution indicating the probability that a trajectory mapping to each cluster is least various distances away from the cluster. Given a new trajectory, a score indicating how unusual the trajectory is may be computed based on the product of the probability of the trajectory mapping to a particular cluster and the intra-cluster probability of the trajectory being a computed distance from the cluster. The distance used to match the trajectory to the cluster and determine intra-cluster probability is computed using a parallel Needleman-Wunsch algorithm, with cells in antidiagonals of a matrix and connected sub-matrices being computed in parallel.

Abstract: Techniques are disclosed for detecting changes in a neuro-linguistic model.