Abstract: A method, non-transitory computer readable medium, and apparatus for side-by-side traffic location load balancing are disclosed. For example, the method receives one or more video images of a side-by-side traffic location, determines a number of cars in a first lane and a number of cars in a second lane of the side-by-side traffic location based upon the one or more video images, calculates a delta between the number of cars in the first lane and the number of cars in the second lane and recommends the first lane or the second lane based upon the delta and a respective weighting factor associated with the first lane and the second lane to provide a load balance of the cars at the side-by-side traffic location.

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: A method, non-transitory computer-readable medium, and apparatus for localizing a region of interest using a dynamic hand gesture are disclosed. For example, the method captures the ego-centric video containing the dynamic hand gesture, analyzes a frame of the ego-centric video to detect pixels that correspond to a fingertip using a hand segmentation algorithm, analyzes temporally one or more frames of the ego-centric video to compute a path of the fingertip in the dynamic hand gesture, localizes the region of interest based on the path of the fingertip in the dynamic hand gesture and performs an action based on an object in the region of interest.

Abstract: A method, non-transitory computer-readable medium, and apparatus for adaptive sampling an ego-centric video to extract features for performing an analysis are disclosed. For example, the method captures the ego-centric video, determines a spatio-temporal location of interest within the ego-centric video, applies an adaptive sampling centered around the spatio-temporal location of interest to obtain one or more spatio-temporal patches, extracts one or more features using the one or more spatio-temporal patches and performs an analysis based on the one or more features.

Abstract: Block-based motion estimation of video compression estimates the direction and magnitude of motion of objects in the scene in a computationally efficient manner and accurately predicts the optimal search direction/neighborhood location for motion vectors, A system can Include a motion detection module that detects apparent motion in the scene, a motion direction and magnitude prediction module that estimates the direction and magnitude of motion of the objects detected to he in motion by the motion detection module, and a block-based, motion estimation module that performs searches in reduced neighborhoods of the target block according to the estimated motion by the motion direction and magnitude prediction module and only for the blocks determined to be in motion by the motion detection module. The Invention is particularly well suited for stationary traffic cameras that monitor roads and highways for traffic law enforcement purposes.

Abstract: A method and system for adaptable video-based object tracking includes acquiring video data from a scene of interest and identifying an initial instance of an object of interest in the acquired video data. A representation of a target object is then established. One or more motion parameters associated with said scene of interest are used to adjust the size of a search neighborhood associated with said target object. The target object is then tracked frame-by-frame in the video data.

Abstract: A system and method for monitoring parking and identifying vehicles by monitoring a parking region based on video data of the parking region received from a video camera, detecting a parking event associated with a vehicle in the parking region, adjusting a view of the video camera based on the parking event, physically tracking the vehicle using the video camera, capturing an image of a license plate of the vehicle, and resuming monitoring the parking region after capturing the image of the license plate.

Abstract: What is disclosed is a system and method for estimating a respiration rate by analyzing distortions in reflections of structured illumination patterns captured in a video containing a view of a subject's thoracic region. In one embodiment, a video of a target region of a body of a subject of interest is received. Video image frames are processed to estimate 3D time-series data for the target region. As more fully disclosed herein, the subject's respiration rate is estimated from the 3D time-series data. Measurements can be acquired under a diverse set of lighting conditions. The teachings hereof provide a non-contact approach to patient respiratory function monitoring that is useful for intensive care units and for monitoring at homes, and which aid in the detection of sudden deterioration of physiological conditions due to changes in respiration rates. The teachings hereof provide an effective tool for non-contact respiratory function study and analysis.

Abstract: A method, non-transitory computer readable medium, and apparatus for training hand detection in an ego-centric video are disclosed. For example, the method prompts a user to provide a hand gesture, captures the ego-centric video containing the hand gesture, analyzes the hand gesture in a frame of the ego-centric video to identify a set of pixels in the image corresponding to a hand region, generates a training set of features from the set of pixels that correspond to the hand region and trains a head-mounted video device to detect the hand in subsequently captured ego-centric video images based on the training set of features.

Abstract: A method, non-transitory computer-readable medium, and apparatus for localizing a region of interest using a dynamic hand gesture are disclosed. For example, the method captures the ego-centric video containing the dynamic hand gesture, analyzes a frame of the ego-centric video to detect pixels that correspond to a fingertip using a hand segmentation algorithm, analyzes temporally one or more frames of the ego-centric video to compute a path of the fingertip in the dynamic hand gesture, localizes the region of interest based on the path of the fingertip in the dynamic hand gesture and performs an action based on an object in the region of interest.

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 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.

Abstract: What is disclosed is a system and method for adaptively reconstructing a depth map of a scene. In one embodiment, upon receiving a mask identifying a region of interest (ROI), a processor changes either a spatial attribute of a pattern of source light projected on the scene by a light modulator which projects an undistorted pattern of light with known spatio-temporal attributes on the scene, or changes an operative resolution of a depth map reconstruction module. A sensing device detects the reflected pattern of light. A depth map of the scene is generated by the depth map reconstruction module by establishing correspondences between spatial attributes in the detected pattern and spatial attributes of the projected undistorted pattern and triangulating the correspondences to characterize differences therebetween. The depth map is such that a spatial resolution in the ROI is higher relative to a spatial resolution of locations not within the ROI.

Abstract: This disclosure provides methods and systems of classifying a vehicle using motion vectors associated with captured images including a vehicle. According to an exemplary method, a cluster of motion vectors representative of a vehicle within a target region is analyzed to determine geometric attributes of the cluster and/or measure a length of a detected vehicle, which provides a basis for classifying the detected vehicle.

Abstract: A method for removing false foreground image content in a foreground detection process performed on a video sequence includes, for each current frame, comparing a feature value of each current pixel against a feature value of a corresponding pixel in a background model. The each current pixel is classified as belonging to one of a candidate foreground image and a background based on the comparing. A first classification image representing the candidate foreground image is generated using the current pixels classified as belonging to the candidate foreground image. The each pixel in the first classification image is classified as belonging to one of a foreground image and a false foreground image using a previously trained classifier. A modified classification image is generated for representing the foreground image using the pixels classified as belonging to the foreground image while the pixels classified as belonging to the false foreground image are removed.

Abstract: The method facilitates efficient motion estimation for video sequences captured with a stationary camera with respect to an object. For video captured with this type of camera, a main cause of changes between adjacent frames corresponds to object motion. In this setting the output from the motion compensation stage is the block matching algorithm describing the way pixel blocks move between adjacent frames. For video captured with cameras mounted on moving vehicles (e.g. school buses, public transportation vehicles and police cars), the motion of the vehicle itself is the largest source of apparent motion in the captured video. In both cases, the encoded set of motion vectors is a good descriptor of apparent motion of objects within the field of view of the camera.

Abstract: A method and system for efficient non-persistent object motion detection comprises evaluating a video segment to identify at least two first pixel classes corresponding to a plurality of stationary pixels and a plurality of pixels in apparent motion, and evaluating the video segment to identify at least two second pixel classes corresponding to a background and a foreground indicative of the presence of a non-persistent object. The first pixel classes and the second pixel classes can be combined to define a final motion mask in the selected video segment indicative of the presence of a non-persistent object. An output can provide an indication that the object is in motion.

Abstract: A method is provided for using a single-pixel imager in order to spatially reconstruct an image of a scene. The method can comprise the following: configuring a light filtering device including an array of imaging elements to a spatially varying optical filtering process of incoming light according to a series of spatial patterns corresponding to sampling functions. The light filtering device can be a transmissive filter including a first membrane, a second membrane, and a variable gap therebetween. The method further comprises tuning a controller for manipulating a variable dimension of the gap; and, measuring, using a photodetector of the single-pixel imager, a magnitude of an intensity of the filtered light across pixel locations in the series of spatial patterns. The magnitude of the intensity can be equivalent to an integral value of the scene across the pixel locations.

Abstract: What is disclosed is system and method for contemporaneously reconstructing images of a scene illuminated with unstructured and structured illumination sources. In one embodiment, the system comprises capturing a first 2D image containing energy reflected from a scene being illuminated by a structured illumination source and a second 2D image containing energy reflected from the scene being illuminated by an unstructured illumination source. A controller effectuates a manipulation of the structured and unstructured illumination sources during capture of the video. A processor is configured to execute machine readable program instructions enabling the controller to manipulate the illumination sources, and for effectuating the contemporaneous reconstruction of a 2D intensity map of the scene using the second 2D image and of a 3D surface map of the scene using the first 2D image. The reconstruction is effectuated by manipulating the illumination sources.

Abstract: What is disclosed is a system and method for estimating tidal chest volume using 3D surface reconstruction based on an analysis of captured reflections of structured illumination patterns from the subject with a video camera. The imaging system hereof captures the reflection of the light patterns from a target area of the subject's thoracic region. The captured information produces a depth map and a volume is estimated from the resulting 3D map. The teachings hereof provide a non-contact approach to patient respiration monitoring that is particularly useful for infant care in a neo-natal intensive care unit (NICU), and can aid in the early detection of sudden deterioration of physiological condition due to detectable changes in respiratory function. The systems and methods disclosed herein provide an effective tool for tidal chest volume study and respiratory function analysis.