Abstract: This disclosure provides a method and system for automatically recognizing facial expressions at variable resolutions of video. According to one exemplary method, facial expressions are detected, extracted and classified from a video sequence based on an automatic localization of the periocular region associated with a detected and extracted face.

Abstract: A method and system for identifying content relevance comprises acquiring video data, mapping the acquired video data to a feature space to obtain a feature representation of the video data, assigning the acquired video data to at least one action class based on the feature representation of the video data, and determining a relevance of the acquired video data.

Abstract: Disclosed is a method and system for detecting an interaction event between two or more objects in a surveillance area, via the application of heuristics to trajectory representations of the static or dynamic movements associated with the objects. According to an exemplary embodiment, trajectory interaction features (TIFs) are extracted from the trajectory representations and heuristics are applied to the TIFs to determine if an interaction event has occurred, such as a potential illegal drug deal involving two or more pedestrians.

Abstract: Disclosed is a method and system for detecting an interaction event between two or more objects in a surveillance area via the application of heuristics to trajectory representations of the static or dynamic movements associated with the objects. According to an exemplary embodiment, trajectory interaction features (TIFs) are extracted from the trajectory representations and heuristics are applied to the TIFs to determine if an interaction event has occurred, such as a potential mugging involving two or more pedestrians.

Abstract: A system and method for detection of drive-arounds in a retail setting. An embodiment includes acquiring images of a retail establishment, analyzing the images to detect entry of a customer onto the premises of the retail establishment, tracking a detected customer's location as the customer traverses the premises of the retail establishment, analyzing the images to detect exit of the detected customer from the premises of the retail establishment, and generating a drive-around notification if the customer does not enter a prescribed area or remain on the premises of the retail location for at least a prescribed minimum period of time.

Abstract: An imaging method includes obtaining an image with a first field of view and first effective resolution and the analyzing the image with a visual attention algorithm to one identify one or more areas of interest in the first field of view. A subsequent image is then obtained for each area of interest with a second field of view and a second effective resolution, the second field of view being smaller than the first field of view and the second effective resolution being greater than the first effective resolution.

Abstract: A method of compressing data in the context of a decision-making task includes receiving raw data, analyzing the raw data to determine content of the raw data, and adjusting one or more one data compression parameters in a compression algorithm. The adjustment of the one or more compression parameters is based on the content of the raw data and a received decision-making task to produce a modified compression algorithm. The raw data is thereafter compressed using the modified compression algorithm and output as compressed data.

Abstract: Disclosed is a method and system for detecting an interaction event between two or more objects in a surveillance area via the application of heuristics to trajectory representations of the static or dynamic movements associated with the objects. According to an exemplary embodiment, trajectory interaction features (TIFs) are extracted from the trajectory representations and heuristics are applied to the TIFs to determine if an interaction event has occurred, such as a potential mugging involving two or more pedestrians.

Abstract: Methods and systems for continuously monitoring the gaze direction of a driver of a vehicle over time. Video is received, which is captured by a camera associated with, for example, a mobile device within a vehicle, the camera and/or mobile device mounted facing the driver of the vehicle. Frames can then be extracted from the video. A facial region can then be detected, which corresponds to the face of the driver within the extracted frames. Features descriptors can then be computed from the facial region. A gaze classifier derived from the vehicle, the driver, and the camera can then be applied, wherein the gaze classifier receives the feature descriptors as inputs and outputs at least one label corresponding to one or more predefined finite number of gaze classes to identify the gaze direction of the driver of the vehicle.

Abstract: A computer vision system (100) operates to monitor an environment (e.g., such as a restaurant, store or other retail establishment) including a resource located therein (e.g., such as a restroom, a dining table, a drink, condiment or supply dispenser, a trash receptacle or a tray collection rack). The system includes: an image source or camera (104) that supplies image data (130) representative of at least a portion of the environment monitored by the system, the portion including the resource therein; and an event detection device (102) including a data processor (112) and operative to detect an event involving the resource. Suitably, the event detection device is arranged to: (i) be selectively configurable by a user to define the event involving the resource; (ii) receive the image data supplied by the image source; (iii) analyze the received image data to detect the defined event; and (iv) output a notification in response to detecting the defined event.

Abstract: A method, non-transitory computer readable medium, and apparatus for directing a vehicle in a side-by-side drive-thru are disclosed. For example, the method receives one or more video images of a side-by-side drive-thru comprising two or more lanes, detects a vehicle approaching an entrance of the side-by-side drive-thru, calculating an estimated order time for the vehicle and directs the vehicle to one of the two or more lanes based on the estimated order time for the vehicle or a previously estimated order time of each one of the a plurality of vehicles already in the first lane and the second lane of the drive-thru.

Abstract: Disclosed is a method and system for detecting an interaction event between two or more objects in a surveillance area, via the application of heuristics to trajectory representations of the static or dynamic movements associated with the objects. According to an exemplary embodiment, trajectory interaction features (TIFs) are extracted from the trajectory representations and heuristics are applied to the TIFs to determine if an interaction event has occurred, such as a potential illegal drug deal involving two or more pedestrians.

Abstract: In a system for detecting location of an object inside of a building, an image capture device of a mobile electronic device captures an image of a boundary of a room in which the portable electronic device is positioned. The system extracts features of a boundary (ceiling, wall or floor) in the image to determine whether the mobile device is in a known location. When the system identifies a known location, it will take an action that provides the portable electronic device with additional functionality at the identified known location. Such functionality may include connecting to a wireless network or communicating with a stationary device at the known location.

Abstract: Methods and systems obtain data representative of a scene across spectral bands using a compressive-sensing-based hyperspectral imaging system comprising optical elements. These methods and systems sample two modes of a three-dimensional tensor corresponding to a hyperspectral representation of the scene using sampling matrices, one for each of the two modes, to generate a modified three-dimensional tensor. After sampling the two modes, such methods and systems sample a third mode of the modified three-dimensional tensor using a third sampling matrix to generate a further modified three-dimensional tensor. Then, the methods and systems reconstruct hyperspectral data from the further modified three-dimensional tensor using the sampling matrices and the third sampling matrix.

Abstract: A system for capturing property damage images includes an imaging device and a knowledge base of property damage incident types. The knowledge base also includes, for each category, one or more image acquisition parameters that are associated with the category. A processor receives a property damage incident type for a property damage claim that is associated with a property that reportedly experienced an incident, retrieves from the knowledge base one or more image acquisition parameters that are associated with the received incident type, and uses the one or more applicable image acquisition parameters to automatically cause the imaging device to capture digital images of the property using the retrieved one or more image acquisition parameters.

Abstract: A method for automatically determining a dynamic queue configuration includes acquiring a series of frames from an image source surveying a queue area. The method includes detecting at least one subject in a frame. The method includes tracking locations of each detected subject across the series of frames. The method includes generating calibrated tracking data by mapping the tracking locations to a predefined coordinate system. The method includes localizing a queue configuration descriptor based on the tracking data.

Abstract: A system segments a set of images of a property to identify a type of damage to the property. The system receives, from an image capturing device, a digital image of a roof or other feature of the property. The system processes the image to identify a set of segments, in which each segment corresponds to a piece of the feature, such as a tab or tooth of a shingle on the roof. The system saves a result of the processing to a data file as a segmented image of the property, and it uses the segmented image to identify a type of damage to the property.

Abstract: A system processes a set of images to verify whether the images depict a type of property damage, such as hail damage from a hail storm. The system receives digital images of a roof or other facet of the property that were taken by an image capturing device after the storm event. The system also receives storm data relating to the storm event and claim data related to a property damage claim. The system uses the claim data to process the digital images to automatically identify damage to the roof or other facet. The system then analyzes the storm data to determine whether the damage is consistent with an actual, naturally-occurring storm event, and it generates a report depicting its findings.

Abstract: A multimodal sensing system includes various devices that work together to automatically classify an action. A video camera captures a sequence of digital images. At least one other sensor device captures other sensed data (e.g., motion data). The system will extract video features from the digital images so that each extracted image feature is associated with a time period. It will extract other features from the other sensed data so that each extracted other feature is associated with a time period. The system will fuse a group of the extracted video features and a group of the extracted other features to create a fused feature representation for a time period. It will then analyze the fused feature representation to identify a class, access a data store of classes and actions to identify an action that is associated with the class, and save the identified action to a memory device.

Abstract: A method, non-transitory computer readable medium, and apparatus for localizing a region of interest using a hand gesture are disclosed. For example, the method acquires an image containing the hand gesture from the ego-centric video, detects pixels that correspond to one or more hands in the image using a hand segmentation algorithm, identifies a hand enclosure in the pixels that are detected within the image, localizes a region of interest based on the hand enclosure and performs an action based on the object in the region of interest.