Abstract: A three-dimensional multilayer skin texture recognition system and method based on hyperspectral imaging. Three-dimensional facial model associated with an object may be acquired from a three-dimensional image capturing device. A face reconstruction approach may be implemented to reconstruct and rewarp the three-dimensional facial model to a frontal face image. A hyperspectral imager may be employed to extract a micro structure skin signature associated with the skin surface. The micro structure skin signature may be characterized utilizing a weighted subtraction of reflectance at different wavelengths that captures different layers under the skin surface via a multilayer skin texture recognition module. The volumetric skin data associated with the face skin can be classified via a volumetric pattern.

Abstract: A multi-camera arrangement for capturing a high resolution image of a target. A first camera may be for capturing a wide field of view low resolution image having a target. The target or a component of it may be border-boxed with a marking. The target may be a human being component, such as a face, having approximately the same size among virtually all humans. A distance of the target may be determined from a known size of a component of the target. The target may be other items of similar size. Coordinates of pixels of the image portion containing the target may be mapped to a pan, tilt and zoom (PTZ) camera. The pan and tilt of the PTZ camera may be adjusted according to image information from the wide field of view camera. Then the PTZ camera may zoom in on the target to obtain a high resolution image of the target.

Abstract: A system for autonomous object tracking with static camera arrangements. Each camera arrangement may minimally have a pan-tilt-zoom camera and a range or depth sensor. Imaging may provide coordinates and depth information of a tracked object. Measurements of an image centroid position and width may be obtained with processing. Maintaining an image at the center of a camera screen may be attained at a pixel width of the image. Estimation and prediction of object size and position may be processed for providing pan, tilt and zoom rates for the camera. Pan, tilt and zoom latency may be accounted for in the system. There may be a number of camera arrangements where tracking of the object may be handed off by one camera arrangement to another.

Abstract: A context-aware smart home energy management (CASHEM) system and method is disclosed. CASHEM dynamically schedules household energy use to reduce energy consumption by identifying contextual information within said household, selecting a comfort of service preference, wherein said comfort of service preference is based on different said contextual information, and extracting an appliance use schedule for maximum energy savings based on said contextual information in light of said comfort of service preferences, by executing a program instruction in a data processing apparatus. CASHEM correlates said contextual information with energy consumption levels to dynamically schedule said appliance based on an energy-saving condition and a user's comfort. Comfort of service preferences are gathered by CASHEM by monitoring occupant activity levels and use of said appliance. CASHEM can also recommend potential energy savings for a user to modify comfort of service preferences.

Abstract: A system having an approach for prioritizing targets for an order of capturing the targets photographically or otherwise. Prioritizing is based on cost of obtaining or capturing the target for viewing or photographing in high resolution. One acquisition mechanism is for obtaining a wide field of view of a scene of targets and another acquisition mechanism is for obtaining a narrow field of view of a target for capture. The cost for prioritizing is based on the time that the narrow field of view acquisition mechanism takes to pan and tilt to get a close-up image of a target divided by the width of the target. The targets may be faces of people.

Abstract: A system and method track a face moving through multiple frames of a video sequence. A predicted position of a face in a video frame is obtained. A search is performed to determine appearance model similarity values of a detected face. Motion and appearance probabilities of the detected face are obtained, and the detected face in the video frame is associated with a track as a function of a joint probabilistic data association filter.

Abstract: Methods and systems for defining, detecting, analyzing, indexing, and retrieving events using video image processing techniques are described herein. An illustrative method of detecting and analyzing events occurring within a region of interest of a video monitoring system may include the steps of extracting information regarding one or more objects located within a region of interest, analyzing the extracted information to detect the occurrence of one or more events within the region of interest, matching at least one of the detected events with a corresponding user-specified event, and performing an action response upon the detection of such an event. In certain embodiments, the detected events may be defined as a set of semantic parameters relating to the object's state and/or physical features, which can be compared against a corresponding set of semantic parameters programmed within a video image-processing appliance manager.

Abstract: A system and method relate to multi-modal biometrics. A single modality score is generated for each of a plurality of biometric modalities. A classifier is selected from a database of multi-modal classifiers, and a multi-modal fusion is applied to the single modality scores using the classifier. The single modality scores are then aggregated. A context dependent model is generated, and a measure of the context in which the biometric samples were obtained is applied to the aggregated single modality scores. It is then determined whether there is a match between two or more biometric samples.

Abstract: Methods and systems for packaging video sequences based on user-specified events are described herein. An illustrative method of event-packaging a video sequence may include the steps of acquiring a video stream containing an event of interest, extracting various event information from the video stream and storing one or more event parameters within an event database, extracting a video clip from the video stream containing the event of interest, associating metadata representing the event parameters to the video clip, and exporting the video clip containing the event of interest and associated metadata to an external agent. In certain embodiments, a video image-processing appliance manager and/or one or more appliance modules can be provided to automatically extract event information from the acquired video stream, and to manage the indexing and storage of event parameters within the event database.

Abstract: A method for automatically estimating the spatial positions between cameras in a camera network utilizes unique identifying signals, such as RFID signals, transmitting between nearby cameras to estimate the relative distances or positions between cameras from received signal strength (RSS), time of arrival (TOA), or time difference of arrival (TDOA) measurements to thereby determine the neighboring relationship among the cameras. A discover-locate process can be used to discover, from the estimated relative distances, unknown cameras in the vicinity of at least three cameras at known locations. Absolute locations of the discovered unknown cameras can then be calculated using a geometric calculation. The discover-locate process can be cascaded throughout the network to discover and locate all unknown cameras automatically using previously discovered and located cameras.

Abstract: A detection system which provides for continuous background estimation removal from a sequence of spectra. A panoramic field of regard may be partitioned into a large number of fields of view (FOVs). An FOV may have a chemical vapor cloud. The small FOV may maximize detection of the cloud. Such detection may require removing the spectral characteristics other than that of the target cloud. This may amount to removal of background spectra with an estimated background developed from one or more FOVs which may or may not be similar to the background of the FOV with the target cloud. A number of estimated background spectra of the other FOVs may be used individually to greatly increase the detection probability of the target chemical.

Abstract: In one embodiment, a method is disclosed that includes obtaining at least one measurement in a spectral domain of a sample and computing one or more measurements of the salient features in the spectral domain. The salient features correspond to at least one peak within the spectral domain. This method also includes classifying the computed salient features against a feature signature of nitric acid. In addition, this method includes determining if the chemical is present in the sample.

Abstract: One embodiment of the application provides a method including segmenting a 3D polygon mesh into a plurality of widgets, defining a state variable for each widget, defining a behavior for each widget, and assembling a three-dimensional user interface from the widgets, the state variables, and the behaviors.

Abstract: Apparatus, systems, and methods for perceiving objects in an environment in three dimensions are provided. One apparatus includes a turntable capable of being coupled to a vehicle and a light detection and ranging (LIDAR) device mounted on the turntable. A system includes a vehicle with a turntable coupled to the vehicle and a LIDAR device mounted on the turntable. One method includes rotating a two-dimensional LIDAR device along an axis of rotation that is substantially normal to a ground plane beneath the vehicle, capturing data points of objects within the environment surrounding the LIDAR device, and generating a three-dimensional representation of the objects based on the data points.

Abstract: A combination of inter-frame and intra-frame and sequential classification methods provide for robust and flexible object classification. Intra-frame classification determines the object type based on features extracted from a single image. Inter-frame classification determines object types based on the features extracted from several sequential frames. A sequential classification combines the results of the inter- and intra-frame classifications over a sequence of images.

Abstract: A system and method are provided for associating faces to determine whether the faces are similar. For example a captured face may be associated with known faces to determine which known face the captured most closely resembles. The system and method incorporate the use of Gabor filters applied to masked face images to develop covariance matrixes. The covariance matrixes are subject to a similarity measure to determine the similarity of the images.

Abstract: A computer-readable medium having computer-executable instructions for performing a method. The method includes determining the transformation of an origin of an imaging device positioned in a vehicle and implementing exponentially stabilizing control laws based on the determined transformation and a distance to an imaged target. The method also includes generating a rotation output from the exponentially stabilizing control laws, generating a zoom output from the exponentially stabilizing control laws, determining a system latency in redirecting an optical axis and modifying a lens system along the optical axis, and determining the transformation of the origin of the imaging device with respect to global coordinates. A centroid of the target image is maintained within a selected distance from the origin of the imaging device. The apparent distance between the imaged target and the imaging device is also maintained.

Abstract: A system and method are provided for tracking a face moving through multiple frames of a video sequence. A predicted position of a face in a video frame is obtained. Similarity matching for both a color model and an edge model are performed to derive correlation values for each about the predicted position. The correlation values are then combined to determine a best position and scale match to track a face in the video.

Abstract: A computer-readable medium having computer-executable instructions for performing a method. The method includes determining the transformation of an origin of an imaging device positioned in a vehicle and implementing exponentially stabilizing control laws based on the determined transformation and a distance to an imaged target. The method also includes generating a rotation output from the exponentially stabilizing control laws, generating a zoom output from the exponentially stabilizing control laws, determining a system latency in redirecting an optical axis and modifying a lens system along the optical axis, and determining the transformation of the origin of the imaging device with respect to global coordinates. A centroid of the target image is maintained within a selected distance from the origin of the imaging device. The apparent distance between the imaged target and the imaging device is also maintained.

Abstract: A method to dynamically stabilize a target image formed on an image plane of an imaging device located in a moving vehicle. The method includes setting an origin in the image plane of the imaging device at an intersection of a first axis, a second axis and a third axis, imaging a target so that an image centroid of the target image is at the origin of the image plane, monitoring sensor data indicative of a motion of the vehicle, and generating pan and tilt output to stabilize the image centroid at the origin in the image plane to compensate for vehicle motion and target motion. The pan and tilt output are generated by implementing exponentially stabilizing control laws. The implementation of the exponentially stabilizing control laws is based at least in part on the sensor data.