Abstract: A data indexing system and method includes acquiring activity data in a context and indexing the activity data in accordance with contextual conditions. The activity data is stored in accordance with indices. An event is correlated with the activity data by using the indices to review the activity data in the context.

Abstract: Methods and systems for processing multilingual DNN acoustic models are described. An example method may include receiving training data that includes a respective training data set for each of two or more or languages. A multilingual deep neural network (DNN) acoustic model may be processed based on the training data. The multilingual DNN acoustic model may include a feedforward neural network having multiple layers of one or more nodes. Each node of a given layer may connect with a respective weight to each node of a subsequent layer, and the multiple layers of one or more nodes may include one or more shared hidden layers of nodes and a language-specific output layer of nodes corresponding to each of the two or more languages. Additionally, weights associated with the multiple layers of one or more nodes of the processed multilingual DNN acoustic model may be stored in a database.

Abstract: Methods and apparatus related to training speech recognition devices are presented. A computing device receives training samples for training a neural network to learn an acoustic speech model. A curriculum function for speech modeling can be determined. For each training sample of the training samples, a corresponding curriculum function value for the training sample can be determined using the curriculum function. The training samples can be ordered based on the corresponding curriculum function values. In some embodiments, the neural network can be trained utilizing the ordered training samples. The trained neural network can receive an input of a second plurality of samples corresponding to human speech, where the second plurality of samples differs from the training samples. In response to receiving the second plurality of samples, the trained neural network can generate a plurality of phones corresponding to the captured human speech.

Abstract: Methods and apparatus related to speech recognition performed by a speech recognition device are disclosed. The speech recognition device can receive a plurality of samples corresponding to an utterance and generate a feature vector z from the plurality of samples. The speech recognition device can select a first frame y0 from the feature vector z, and can generate a second frame y1, where y0 and y1 differ. The speech recognition device can generate a modified frame x? based on the first frame y0 and the second frame y1 and then recognize speech related to the utterance based on the modified frame x?. The recognized speech can be output by the speech recognition device.

Abstract: A system and method for presentation of surveillance data includes a first environment including one or more sensors configured to monitor the first environment wherein data is collected by the one or more sensors. A second environment includes one or more information presentation devices configured to convey an event or state occurring in the first environment by altering an ambient sensory condition in the second environment, such that the ambient sensory condition is modified in accordance with a type and/or level of the event or state.

Abstract: A system and method for presentation of surveillance data includes a first environment including one or more sensors configured to monitor the first environment wherein data is collected by the one or more sensors. A second environment includes one or more information presentation devices configured to convey an event or state occurring in the first environment by altering an ambient sensory condition in the second environment, such that the ambient sensory condition is modified in accordance with a type and/or level of the event or state.

Abstract: A method and system for is disclosed for speech synthesis using deep neural networks. A neural network may be trained to map input phonetic transcriptions of training-time text strings into sequences of acoustic feature vectors, which yield predefined speech waveforms when processed by a signal generation module. The training-time text strings may correspond to written transcriptions of speech carried in the predefined speech waveforms. Subsequent to training, a run-time text string may be translated to a run-time phonetic transcription, which may include a run-time sequence of phonetic-context descriptors, each of which contains a phonetic speech unit, data indicating phonetic context, and data indicating time duration of the respective phonetic speech unit. The trained neural network may then map the run-time sequence of the phonetic-context descriptors to run-time predicted feature vectors, which may in turn be translated into synthesized speech by the signal generation module.

Abstract: There is provided an apparatus for the certification of privacy compliance. The apparatus includes a registry of at least one of enrolled video surveillance operators, approved surveillance hardware devices, approved surveillance software programs, approved surveillance system installers, and approved entities that manage surveillance systems. The apparatus further includes a registry searcher, in signal communication with the registry, for receiving queries to the registry, and for determining whether at least one of a particular surveillance operator, a particular surveillance hardware device, a particular surveillance software program, a particular surveillance system installer, and a particular entity that manages a particular surveillance system is on the registry based on a given query.

Abstract: A system for automatically acquiring high-resolution images by steering a pan-tilt-zoom camera at targets detected in a fixed camera view is provided. The system uses automatic or manual calibration between multiple cameras. Using automatic calibration, the homography between the cameras in a home position is estimated together with the effects of pan and tilt controls and the expected height of a person in the image. These calibrations are chained together to steer a slave camera. The manual calibration scheme steers a camera to the desired region of interest and calculates the pan, tile and zoom parameters accordingly.

Abstract: A system and method for presentation of surveillance data includes a first environment including one or more sensors configured to monitor the first environment wherein data is collected by the one or more sensors. A second environment includes one or more information presentation devices configured to convey an event or state occurring in the first environment by altering an ambient sensory condition in the second environment, such that the ambient sensory condition is modified in accordance with a type and/or level of the event or state.

Abstract: A system for automatically acquiring high-resolution images by steering a pan-tilt-zoom camera at targets detected in a fixed camera view is provided. The system uses automatic or manual calibration between multiple cameras. Using automatic calibration, the homography between the cameras in a home position is estimated together with the effects of pan and tilt controls and the expected height of a person in the image. These calibrations are chained together to steer a slave camera. The manual calibration scheme steers a camera to the desired region of interest and calculates the pan, tile and zoom parameters accordingly.

Abstract: There is provided an apparatus for the certification of privacy compliance. The apparatus includes a registry of at least one of enrolled video surveillance operators, approved surveillance hardware devices, approved surveillance software programs, approved surveillance system installers, and approved entities that manage surveillance systems. The apparatus further includes a registry searcher, in signal communication with the registry, for receiving queries to the registry, and for determining whether at least one of a particular surveillance operator, a particular surveillance hardware device, a particular surveillance software program, a particular surveillance system installer, and a particular entity that manages a particular surveillance system is on the registry based on a given query.

Abstract: A system for automatically acquiring high-resolution images by steering a pan-tilt-zoom camera at targets detected in a fixed camera view is provided. The system uses automatic or manual calibration between multiple cameras. Using automatic calibration, the homography between the cameras in a home position is estimated together with the effects of pan and tilt controls and the expected height of a person in the image. These calibrations are chained together to steer a slave camera. The manual calibration scheme steers a camera to the desired region of interest and calculates the pan, tile and zoom parameters accordingly.

Abstract: Systems and methods for foreground analysis in real-time video include background subtraction and foreground detection, shadow removal, quick lighting change adaptation, static foreground region detection, foreground fragment reduction, and frame level change detection. Processes include background image extraction and foreground detection, integrating texture information of the background image and a current frame to remove false positive foreground areas resulting from lighting changes, integrating pixel intensity information by determining a cross-correlation of intensities between a current frame and the background image for each pixel in a foreground mask to remove image shadows. Static foreground region detection and fragment reduction are also included.

Abstract: Techniques for tracking entities using a single overhead camera are provided. A foreground region is detected in a video frame of the single overhead camera corresponding to one or more entities. It is determined if the foreground region is associated with an existing tracker. It is determined whether the detected foreground region is the result of at least one of a merger of two or more smaller foreground regions having corresponding existing trackers and a split of a larger foreground region having a corresponding existing tracker when the detected foreground region is not associated with an existing tracker. The detected foreground region is tracked via at least one existing tracker when the foreground region is associated with an existing tracker or the foreground region is the result of at least one of a merger and a split.

Abstract: Systems and methods for foreground analysis in real-time video include background subtraction and foreground detection, shadow removal, quick lighting change adaptation, static foreground region detection, foreground fragment reduction, and frame level change detection. Processes include background image extraction and foreground detection, integrating texture information of the background image and a current frame to remove false positive foreground areas resulting from lighting changes, integrating pixel intensity information by determining a cross-correlation of intensities between a current frame and the background image for each pixel in a foreground mask to remove image shadows. Static foreground region detection and fragment reduction are also included.

Abstract: Methods and apparatus are provided for normalizing objects across a plurality of image viewpoints. A set of classification results are obtained for a given object class across a sequence of images for each of a plurality of viewpoints. The classification results are each comprised of a position of one of the objects in the image, and at least one projected property of the object at that position. Normalization parameters are then determined for each of the viewpoints by fitting a high order model to the classification results to model a change in the projected property. The high order model may implement a least squares fit of a second order polynomial to the classification results. The normalization parameters may be used to compute normalized features and normalized training data for object classification.

Abstract: A surveillance system and method includes a plurality of sensors configured to monitor an environment. A plurality of analytic engines is associated with each of the plurality of sensors. The plurality of analytic engines employs different technologies and is configured to analyze input from the sensors to determine whether an event has occurred in a respective technology. A unifying data model is configured to cross correlate detected events from the different technologies to gain integrated situation awareness across the different technologies.

Abstract: A system and method for generating a viewable video index for low bandwidth applications are provided. The exemplary aspects of the present invention solve the problems with the prior art systems by incorporating information for generating a viewable representation of the video data into the index, thus generating a viewable video index. The viewable video index contains information for generating a visual representation of moving objects in the video data, a visual representation of the background of the video capture area, i.e. the scene, a representation of the object trajectory, a representation of the object attributes, and a representation of detected events. The result is that the viewable video index may be transmitted to a low bandwidth application on a client device and may be used along with associated object and background models to generate a representation of the actual video data without requiring that the original video data itself be streamed to the client device.

Abstract: A system for automatically acquiring high-resolution images by steering a pan-tilt-zoom camera at targets detected in a fixed camera view is provided. The system uses automatic or manual calibration between multiple cameras. Using automatic calibration, the homography between the cameras in a home position is estimated together with the effects of pan and tilt controls and the expected height of a person in the image. These calibrations are chained together to steer a slave camera. The manual calibration scheme steers a camera to the desired region of interest and calculates the pan, tile and zoom parameters accordingly.