Abstract: Methods are apparatuses are described for identifying a target object using optical occlusion. A head-mounted display perceives a characteristic of a reference object. The head-mounted display detects a change of the perceived characteristic of the reference object and makes a determination that a detected object caused the change of the perceived characteristic. In response to making the determination, the head-mounted display identifies the detected object as the target object.

Abstract: An arrival time estimation device includes an image input unit configured to input an image signal to each frame, an object detecting unit configured to detect an object indicated by the image signal input through the image input unit, and an arrival time calculating unit configured to calculate a rotation matrix indicating rotation of an optical axis of an imaging device that captures the image signal based on a direction vector indicating a direction to the object detected by the object detecting unit, to calculate a change in a distance to the object based on a vector obtained by multiplying a past direction vector by the calculated rotation matrix and a current direction vector, and to calculate an arrival time to the object based on the calculated distance change.

Abstract: An eye state detection apparatus includes a camera, a first calculator, a memory, a second calculator, and a third calculator. The camera obtains a plurality of face images of a driver. The first calculator calculates an opening amount of an eye of the driver based on each face image. The memory stores the opening amounts calculated by the first calculator. The second calculator groups the opening amounts into a plurality of groups in a sequential manner, calculates a group distribution of each group, calculates an entire distribution of all of the opening amounts, and sets the entire distribution as a reference distribution when a difference among the group distributions is within a predetermined range. The third calculator calculates an opening degree of the eye based on the reference distribution of the opening amounts when the reference distribution of the opening amounts is calculated by the second calculator.

Abstract: Methods and apparatus are described for monocular 3D human pose estimation and tracking, which are able to recover poses of people in realistic street conditions captured using a monocular, potentially moving camera. Embodiments of the present invention provide a three-stage process involving estimating (10, 60, 110) a 3D pose of each of the multiple objects using an output of 2D tracking-by detection (50) and 2D viewpoint estimation (46). The present invention provides a sound Bayesian formulation to address the above problems. The present invention can provide articulated 3D tracking in realistic street conditions. The present invention provides methods and apparatus for people detection and 2D pose estimation combined with a dynamic motion prior. The present invention provides not only 2D pose estimation for people in side views, it goes beyond this by estimating poses in 3D from multiple viewpoints. The estimation of poses is done in monocular images, and does not require stereo images.

Abstract: A technique is provided for recognizing faces in an image stream using a digital image acquisition device. A first acquired image is received from an image stream. A first face region is detected within the first acquired image having a given size and a respective location within the first acquired image. First faceprint data uniquely identifying the first face region are extracted along with first peripheral region data around the first face region. The first faceprint and peripheral region data are stored, and the first peripheral region data are associated with the first face region. The first face region is tracked until a face lock is lost. A second face region is detected within a second acquired image from the image stream. Second peripheral region data around the second face region are extracted. The second face region is identified upon matching the first and second peripheral region data.

Abstract: An information processing device includes: an outline extraction unit extracting an outline of a subject from a picked-up image of the subject; a characteristic amount extraction unit extracting a characteristic amount, by extracting sample points from points making up the outline, for each of the sample points; an estimation unit estimating a posture of a high degree of matching as a posture of the subject by calculating a degree of the characteristic amount extracted in the characteristic amount extraction unit being matched with each of a plurality of characteristic amounts that are prepared in advance and represent predetermined postures different from each other; and a determination unit determining accuracy of estimation by the estimation unit using a matching cost when the estimation unit carries out the estimation.

Abstract: Techniques for ability enhancement are described. Some embodiments provide an ability enhancement facilitator system (“AEFS”) configured to enhance a user's ability to operate or function in a transportation-related context as a pedestrian or a vehicle operator. In one embodiment, the AEFS is configured perform vehicular threat detection based at least in part on analyzing image data. An example AEFS receives data that represents an image of a vehicle. The AEFS analyzes the received data to determine vehicular threat information, such as that the vehicle may collide with the user. The AEFS then informs the user of the determined vehicular threat information, such as by transmitting a warning to a wearable device configured to present the warning to the user.

Abstract: A system and method for performing a depth estimation procedure utilizing defocused pillbox images includes a camera device with a sensor device for capturing pillbox blur images of a photographic target. The camera utilizes a depth estimator for performing a Gaussianization procedure that transforms the pillbox blur images into corresponding Gaussian blur images. The Gaussianization procedure is performed by convolving the pillbox blur images with a Gaussianization kernel to generate the corresponding Gaussian blur images. The depth estimator then utilizes the Gaussian blur images for effectively performing the depth estimation procedure.

Abstract: A distance measurement apparatus and a distance measurement method are provided. The apparatus includes a line-shaped laser transmitter, an image sensing device and a processing unit. The line-shaped laser transmitter transmits a line-shaped laser, and the image sensing device senses the line-shaped laser to output a line-shaped laser image. The processing unit receives the line-shaped laser image, and segments the line-shaped laser image into several sub-line-shaped laser images. The processing unit further calculates a vertical position for a laser line in each sub-line-shaped laser image, and outputs each distance information according to the corresponding sub-line-shaped laser image and a transformation relation.

Abstract: Disclosed is method of measuring the displacement between a reference region of a reference image and a sample region of a sample image. The method spatially varies the reference region using a one-dimensional filter having complex kernel values, wherein a length (radius) and direction (angle or tangent segment) of the filter is a function of position in the reference region. The method then measures a displacement between the reference region and the sample region by comparing the spatially varied reference region and the sample region.

Abstract: The invention relates to a method which uses a series of images sensed by an image sensor including at least one preceding image and one following image to estimate movement. A first estimated movement is initially obtained by estimating the total movement from the preceding image to the following image. Next, an image compensated according to the first estimated movement is obtained from either one of the preceding and following images. Then, a second estimated movement is obtained by estimating dense movement between the compensated image and the other from the preceding and following images. Next, a residual value of global movement is determined. Finally, if the residual value is lower than a threshold value the second estimated movement is provided; otherwise, the preceding steps are repeated.

Abstract: A server system stores a database of records. Each record stores a digital image of a pet, facial recognition markers, and owner identification information. An owner may register his/her pet using a cellular telephone and a mobile app. The owner is prompted to supply a digital photograph of the pet and to enter eye and nose location information. The system generates therefrom facial recognition markers and stores the markers in a record. If a lost pet is found, a finding user can use a cellular telephone to take a picture of the animal and to enter eye and nose location information. The system generates markers and compares the markers in a facial recognition process to other stored markers to identify records of animals that resemble the lost pet. If the record of the lost pet is identified, then owner identification information in the record is used to contact the owner.

Abstract: Face recognition using multilayered discriminant analysis includes systems and methods applying an initial linear discriminate analysis to a database of face images in a more-or less conventional manner. Initial fuzzy logic then is applied to the results of the initial linear discriminate analysis to produce a subset of the database of face images. Thereafter, a subsequent linear discriminate analysis is applied to the subset of the database of face images and subsequent fuzzy logic is applied to the results of the subsequent linear discriminate analysis to produce a further subset of the subset of the database of face images. The application of the subsequent linear discriminate analysis and application of the subsequent fuzzy logic may be repeated until the further subset contains only one, or zero, face images.

Abstract: A medical claim fraud prevention system includes a first controller positioned at a medical facility and a second controller positioned at a patient information collection center and adapted to be in communication with the first controller. The first controller may transmit a first signal to the second controller responsive to an occurrence of an event. The second controller may initiate contact with a photograph database via a network interface in response to the first signal received from the first controller, and an official photograph associated with the patient name may be identified. The second controller may transmit the official photograph to the first controller so that the official photograph is viewable on a user interface at the medical facility. A patient identity may be verified by performing a comparison of a patient with the official photograph of the patient.

Abstract: The image processing apparatus detects a region of a subject from an input image, and extracts an image feature amount from the region. Also, the apparatus classifies the subject into any one of plural attributes based on the image feature amount, and estimates, based on the image feature amount and an attribute into which it is classified, an attribute value of the subject belonging to the attribute into which it is classified.

Abstract: A system and method for tagging an image of an individual in a plurality of photos is disclosed herein. A feature vector of an individual is used to analyze a set of photos on a social networking website such as www.facebook.com to determine if an image of the individual is present in a photo of the set of photos. Photos having an image of the individual are tagged preferably by listing a URL or URI for each of the photos in a database.

Abstract: The present invention provides a fingerprint recognition apparatus and a fingerprint recognition method adapted for the apparatus. The method includes steps of: performing an image processing function for fingerprints of a user and generating an image formed by a number of ridges; scanning the image according to a scanning frequency and selecting a number of scanning lines; acquiring a cross point formed between each ridge of fingerprint and each scanning line; acquiring a tangent line; generating a graph between each cross point and each corresponding slope value; determining whether the graph is similar to one of the stored at least one graph to obtain a determination result; and outputting a recognition result of the fingerprints of the user associated with the determination result.

Abstract: Enhanced biometric authentication is achieved by combining a user's inherent biometric data with the user's knowledge of a secret glyph. In one embodiment, a touchpad is provided on which the user may use a finger to indicate a plurality of strokes that form a distinct glyph. Image stabilization may be used to extract a readable fingerprint from the strokes, and the glyph and finger print are matched to a stored profile. The glyph may be one or more alphanumeric characters that represent a password. The user can then enter the password on the touch pad with his finger. If the fingerprint and password both match, the user is authenticated.

Abstract: A biometric authentication device including: a biometric information acquiring unit which generates a biometric input image representing user's biometric input information; and a processing unit.

Abstract: A method of processing graphical image data representing optically scanned medication-related units may include receiving image data generated responsive to disposal of the units on a tray disposed a distance from an image acquisition component, the image data including data indicative of visually observable features of the units disposed on the tray. The method further includes comparing at least two features among the visually observable features from the image data to reference data indicative of corresponding features of reference units. The reference data is selected for comparison based on an identification of the reference data as corresponding to a prescription being processed. The reference data includes data indicative of features of the reference units extracted from images captured using hardware corresponding to hardware used to generate the image data. The method further includes generating a likelihood rating for each of the at least two features based on the comparing.

Abstract: A method for modifying a first data set, which is called the target data set and represents the contour of an anatomical object, the method comprising the steps of: matching a second data set, which is called the template data set and comprises a template for the modification and a template of the object, to the target data set such that the object template in the template data set matches the object in the target data set, thereby also adapting the modification template; applying the adapted modification template to the target data set.

Abstract: The invention provides methods and systems for reconstructing feature intensities from pixel level data. In certain embodiments, the invention uses an empirically determined transfer function to construct a theoretical estimate of pixel level data and then iteratively updates feature intensities based on a minimum multiplicative error between the pixel level data and the theoretical estimate of the pixel level data.

Abstract: Compressed sensing (CS) estimation approaches rely on a priori sparsity to significantly reduce the number of samples needed to provide high sampling fidelity, relative to the normal Shannon-Nyquist limit. Accordingly, CS approaches are of considerable interest for detector multiplexing in applications which have inherently sparse signals (e.g., the two correlated photon detection events in PET imaging). However, CS approaches also tend to fare poorly in the presence of noise, which has limited their applicability in practice. In this work, we show that CS estimation can be used to provide an estimate of the support of an image. This estimated support is then used as a constraint for maximum likelihood image reconstruction. This approach has robust noise performance and provides high reconstruction fidelity.

Abstract: An apparatus comprises: an imaging system (10) configured to acquire emission data from a cyclically varying element; a monitoring instrument (20, 22) configured to measure the cyclical varying of the cyclically varying element; and an electronic device (40) configured to locate an image feature corresponding to the cyclically varying element in the acquired emission data based on correlation of time variation of the emission data with the cyclical varying of the cyclically varying element measured by the monitoring instrument.

Abstract: An anatomical image data set and an emission image data set are acquired for a subject. An attenuation map is generated from the anatomical image data set. The emission image data set is reconstructed to generate an emission image. The reconstructing includes correcting for attenuation of emission radiation in the subject using the attenuation map. A value is calculated for a quality assurance (QA) metric quantifying alignment of the attenuation map with the emission image. The emission image is displayed or printed together with the calculated quality assurance metric. In some embodiments, prior to the reconstructing the attenuation map is registered with the emission image data set by performing a global rigid registration followed by a local non-rigid registration of a region of interest.

Abstract: A method of processing an image is provided. The method includes generating a tissue emphasis image by emphasizing a predetermined tissue of at least two radiation images of different energy bands, and generating a diagnostic image by combining at least one of the at least two radiation images of different energy bands and the tissue emphasis image.

Abstract: In accordance with preferred embodiments of the present invention, a method for imaging tissue, for example, includes the steps of mounting the tissue on a computer controlled stage of a microscope, determining volumetric imaging parameters, directing at least two photons into a region of interest, scanning the region of interest across a portion of the tissue, imaging a plurality of layers of the tissue in a plurality of volumes of the tissue in the region of interest, sectioning the portion of the tissue and imaging a second plurality of layers of the tissue in a second plurality of volumes of the tissue in the region of interest, detecting a fluorescence image of the tissue due to said excitation light; and processing three-dimensional data that is collected to create a three-dimensional image of the region of interest.

Abstract: An exemplary aspect of the present disclosure, there is provided a light detection device including a first detecting portion to detect fluorescence from a first detection position on a conveyor path along which a sheet is conveyed, a second detecting portion to detect afterglow from a second detection position on the conveyor path along which the sheet is conveyed, a lighting portion to apply excitation light over a range including the first detection position and not including the second detection position, a first reference member which includes fluorescent material that emits fluorescence by being excited by the lighting portion and which emits reference light toward the first detecting portion, a second reference member to emit reference light toward the second detecting portion when the sheet is present at the second detection position, the reference light not influencing a result of detection by the second detecting portion, and a correction controller to correct a result of detection by the first dete

Abstract: A system and method for generating robust depth maps includes a depth estimator that creates a depth map pyramid structure that includes a plurality of depth map levels that each have different resolution characteristics. In one embodiment, the depth map levels include a fine-scale depth map, a medium-scale depth map, and a coarse scale depth map. The depth estimator evaluates depth values from the fine-scale depth map by utilizing fine-scale confidence features, and evaluates depth values from the medium-scale depth map and the coarse-scale depth map by utilizing coarse-scale confidence features. The depth estimator then fuses optimal depth values from the different depth map levels into an optimal depth map.

Abstract: A detection device capable of reliably detecting an object to be detected. An intersection region pattern setting unit (106) sets a configuration pattern of a first intersection region pattern group in sequence for each unit image pair. Each intersection region pattern is defined by set image information which denotes locations and sizes of regions (where n is a natural number greater than 1) within respective unit images (e.g., unit image plane coordinates), as well as whether each region is set within either or both of a first unit image and a second unit image. A detection unit (108) detects the object to be detected, based on a total feature value relating to each configuration pattern of the first intersection region pattern group, computed by a feature value computation unit (107), and a strong identification apparatus configured from a plurality of weak identification apparatuses and stored in an identification apparatus storage unit (112).

Abstract: A system and method for creating a gesture and generating a classifier that can identify the gesture for use with an application is described. The designer constructs a training set of data containing positive and negative examples of the gesture. Machine learning algorithms are used to compute the optimal classification of the training data into positive and negative instances of the gesture. The machine learning algorithms generate a classifier which, given input data, makes a decision on whether the gesture was performed in the input data or not.

Abstract: Techniques are described herein for selecting representative images for video items using a trained machine learning engine. A training set is fed to a machine learning engine. The training set includes, for each image in the training set, input parameter values and an externally-generated score. Once a machine learning model has been generated based on the training set, input parameters for unscored images are fed to the trained machine learning engine. Based on the machine learning model, the trained machine learning engine generates scores for the images. To select a representative image for a particular video item, candidate images for that particular video item may be ranked based on their scores, and the candidate image with the top score may be selected as the representative image for the video item.

Abstract: An optimized adaptive loop filter does not redesign filters inside the optimization loop of signaling depth which saves computations. Additionally, the Sum of Squared Errors (SSE) (distortion) of blocks is computed for the smallest blocks, thus, allowing for the distortion of larger blocks to be computed efficiently by adding block SSEs together which saves computations by removing redundant operations to calculate SSE of a block each time.

Abstract: An image recognition information attaching apparatus includes a retrieving unit that retrieves image information on a per piece basis of identification information, from the image information having the identification information associated thereto in advance, a generator unit that generates feature information from the image information retrieved by the retrieving unit, and a learning unit that provides a learning result by learning a relation between the feature information generated by the generator unit and the identification information of the image information corresponding to the feature information, using a stochastic model including a mixture of a plurality of probability distributions.

Abstract: A method and system for probe-based pattern matching including an apparatus for synthetic training of a model of a pattern. The apparatus comprises a sensor for obtaining an image of the pattern and a processor for receiving the image of the pattern from the sensor and running a program. In the steps performed by the program a boundary of the pattern in the image is identified. A plurality of positive probes are placed at selected points along the boundary of the pattern and at least one straight segment of the boundary of the pattern is identified. The at least one straight segment of the boundary is extended to provide an imaginary straight segment and a plurality of negative probes are placed at selected points along the imaginary straight segment, where each negative probe has a negative weight.

Abstract: Included are (a) performing processes on second training data items stored in a training database to generate third training data items each obtained through a corresponding one of the processes, (b) selecting, from among the third training data items generated in step (a), a selection data item having a highest similarity to a feature data item of the input image, (c) generating a high-frequency data item by: determining (i) the second training data item used in generating the selection data item and (ii) a first process performed on the second training data item to generate the selection data item; and performing the first process on the first training data item that is paired with the determined second training data item; and (d) generating an output image by adding an image indicated by the high-frequency data item to the input image.

Abstract: The disclosure relates to a system and a method for generating clothing feature data representative of at least one clothing feature of a piece of clothing being worn by the person in a set of images, and training a discriminative clothing classifier using the clothing feature data to provide a personal clothing model that corresponds to the piece of clothing. The personal clothing model can be used to identify additional images in which the person appears.

Abstract: Disclosed is a distortion invariant system, method and computer readable medium for detecting the presence of one or more predefined targets in an input image. The input image and a synthetic discriminant function (SDF) reference image are correlated in a shift phase-encoded fringe-adjusted joint transform correlation (SPFJTC) correlator yielding a correlation output. A peak-to-clutter ratio (PCR) is determined for the correlation output and compared to a threshold value. A predefined target is present in the input image when the PCR is greater than or equal to the threshold value.

Abstract: A system and a method of detecting abnormal or security-significant segments of a video are provided. The video is divided into several segments. A set of features of each segment of the video is extracted in order to calculate a set of factors corresponding to each segment. A value deemed abnormal of each segment is calculated according to the set of factors corresponding to each segment. One or more abnormal segments are determined from the segments based on the abnormal values. The set of features includes a color variable feature, a movement variable feature, a movement variable ratio feature and a background variable feature.

Abstract: An image recognition apparatus comprises an image obtainment unit configured to obtain an image; a region setting unit configured to set at least one local region in the image; a feature extraction unit configured to extract feature patterns from the local region; a setting unit configured to set, out of a plurality of bins corresponding to a plurality of patterns which can form the feature patterns, bins that have been predetermined in accordance with a type of the local region as histogram bins used in generating a histogram; a generation unit configured to generate a histogram corresponding to the extracted feature patterns using the set histogram bins; and a recognition unit configured to perform image recognition using the generated histogram.

Abstract: In various embodiments, methods and systems are disclosed for dynamic runtime implementation and end-to-end biased tuning of a two stage image classification system based on a decision function that uses network packet sizes and multiple image characteristics to determine the selection of an encoding codec to reduce overall network bandwidth consumption.

Abstract: Apparatus, systems and methods for low latency remote display rendering using tile-based rendering systems are disclosed. In one implementation, a system includes a network interface and a content source coupled to the network interface. The content source being capable of rendering at least one tile of a tiled image, encoding the at least one tile, and providing the at least one encoded tile to the network interface before all tiles of the tiled image are rendered.

Abstract: A color processing apparatus includes the following elements. A point group generator generates a point group existing in a color gamut of an output apparatus. A design point group extracting unit extracts, from the point group, a design point group that satisfies predetermined design conditions. A corresponding point group setting unit sets, on the basis of the design point group, a corresponding point group indicating a boundary of a color gamut represented by the design point group, such that points of the corresponding point group are associated with individual points of a boundary point group existing on a boundary of the color gamut of the output apparatus. A design color gamut boundary generator generates a boundary of a design color gamut that satisfies the predetermined design conditions, by using a set of the boundary point group and the corresponding point group.

Abstract: An information processing apparatus encodes an input pattern to a code including a plurality of bits, calculates reliabilities for respective bits of the code, generates a similar codes each similar to the code based on the reliabilities, and recognizes the input pattern based on the code and the similar codes.

Abstract: A method for aligning and unwarping distorted images in which lens profiles for a variety of lens and camera combinations are precomputed. Metadata stored with images is used to automatically determine if a set of component images include an excessive amount of distortion, and if so the metadata is used to determine an appropriate lens profile and initial unwarping function. The initial unwarping function is applied to the coordinates of feature points of the component images to generate substantially rectilinear feature points, which are used to estimate focal lengths, centers, and relative rotations for pairs of the images. A global nonlinear optimization is applied to the initial unwarping function(s) and the relative rotations to generate optimized unwarping functions and rotations for the component images. The optimized unwarping functions and rotations may be used to render a panoramic image.

Abstract: A method of tracking targets in video data. At each of a sequence of time steps, a set of weighted probability distribution components is derived. At each time step the following steps are performed. First, a new set of components from the components of the previous time step are derived in accordance with a predefined motion model for the targets. The video at the current time step is then analysed to obtain a set of measurements, and the new set of components is updated using the measurements in accordance with a predefined measurement model. Finally, the set of components derived at each time step are analysed to derive a set of tracks for the targets.

Abstract: The present disclosure relates to a system and a method for finger printing for comics. A system for searching comics according to the present disclosure includes: a fingerprint database storing fingerprints extracted from comics, a comics fingerprint extraction unit extracting fingerprints configured of at least one of box frames, cuts, and speech bubbles included in input comic images, a fingerprint based candidate group search unit searching candidate groups among comics stored in the fingerprint database using the extracted fingerprints, and a similarity measuring unit measuring similarity between the searched candidate groups and the comic images corresponding to the extracted fingerprints.

Abstract: An information processing apparatus sets a plurality of reference locations of data in information as one reference location pattern and acquires a feature amount obtained from a value of data of one of the plurality of pieces of reference information in one reference location pattern for each of a plurality of reference location patterns and the plurality of pieces of reference information. The apparatus extracts data included in the input information according to each of the plurality of reference location patterns, selects the reference location pattern for classification of the input information from the plurality of reference location patterns based on a value of data included in the extracted input information, and executes classification of the input information by using the feature amount in the selected reference location pattern and data included in the input information at a reference location indicated by the reference location pattern.

Abstract: A database of user preferences for a high quality picture is maintained. Preferences may be generated over time by tracking attributes of pictures that the user has deleted or failed to select for storage. When the camera is in preview mode, the camera may automatically capture image data for one or more pictures as a background operation. When the user commands the taking of a picture, the camera will capture image data for a user photo. A comparison may be made to determine which one of the automatically taken picture or the user photo has a higher correlation to user preferences for a quality photo. If the user photo has higher correlation, it may be retained in the customary manner for digital pictures. If the automatically taken photo has higher correlation, the user may be given the option to select the user photo or the automatically taken photo for retention.

Abstract: In a pose estimation for estimating the pose of an object of pose estimation with respect to a reference surface that serves as a reference for estimating a pose, a data processing device: extracts pose parameters from a binarized image; identifies a combination of pose parameters for which the number of cross surfaces of parameter surfaces that accord with surface parameter formulas, which are numerical formulas for expressing a reference surface, is a maximum; finds a slope weighting for each of cross pixels, which are pixels on each candidate surface and which are pixels within a prescribed range, that is identified based on the angles of the tangent plane at the cross pixel and based on planes formed by each of the axes of parameter space; and identifies the significant candidate surface for which a number, which is the sum of slope weightings, is a maximum, as the actual surface that is the reference surface that actually exists in the image.