Abstract: A gaze estimation method and apparatus is disclosed. The gaze estimation method includes obtaining an image including an eye region of a user, extracting, from the obtained image, a first feature of data, obtaining a second feature of data used for calibration of a neural network model, and estimating a gaze of the user using the first feature and the second feature.

Abstract: Systems and methods for eye image segmentation and image quality estimation are disclosed. In one aspect, after receiving an eye image, a device such as an augmented reality device can process the eye image using a convolutional neural network with a merged architecture to generate both a segmented eye image and a quality estimation of the eye image. The segmented eye image can include a background region, a sclera region, an iris region, or a pupil region. In another aspect, a convolutional neural network with a merged architecture can be trained for eye image segmentation and image quality estimation. In yet another aspect, the device can use the segmented eye image to determine eye contours such as a pupil contour and an iris contour. The device can use the eye contours to create a polar image of the iris region for computing an iris code or biometric authentication.

Abstract: Systems, methods, and computer-readable media for facilitating an authentication processing service are provided.

Abstract: Methods, devices, and systems are provided for providing continuous authentication to a user having a wearable device in an access control system. The wearable device allows the authentication of the user to be maintained until an authentication interruption signal is received. The interruption signal may be based on user biometrics, a state of the wearable device, a communication range of the wearable device with a trusted mobile device, and more. Upon receiving the interruption signal, the continuous authentication for the wearable device, and the user, may be revoked, destroyed, or disabled.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium for red eye detection are provided. In one aspect, a system includes an image acquisition device, first and second illuminators, and at least one processor. The first illuminator is arranged closer to the image acquisition device than the second illuminator. The image acquisition device is configured to capture a first facial image of a face of a subject with the first illuminator being on and the second illuminator being off and a second facial image of the face of the subject with the second illuminator being on and the first illuminator being off. The processor can process the first facial image based on the second facial image to determine whether at least one eye of the subject is live by determining that the first facial image includes a red eye reflection from the at least one eye.

Abstract: To enable a more accurate estimation of the center of a pupil even in a situation in which it is difficult to capture an image of the entire pupil. An information processing device includes: an acquisition unit that acquires information regarding an iris in an eyeball; and an estimation unit that estimates a center position of a pupil in the eyeball on the basis of the information regarding the iris.

Abstract: Provided are methods, apparatuses, and systems for tracking a pupil of an eye of a user with degraded iris authentication accuracy. In certain examples, an eye tracking device includes a camera adapted to capture an eye image of at least one eye of a user. The eye tracking device further includes an image processor adapted to defocus the eye image in order to generate a defocused eye image. The defocused eye image comprises reduced iris authentication accuracy. The eye tracking device further includes a tracking processor configured to detect a pupil in the defocused eye image and determine a gaze direction of the user based at least in part on one or more of the defocused eye image and the pupil.

Abstract: A mobile terminal device includes: a storage device that stores shot images taken by a camera; a face recognizer that recognizes a person's face image contained in a shot image selected from among the shot images stored in the storage device; a determiner that determines whether or not the person's face image recognized by the face recognizer is a camera-eye-aligned image where an eye direction of the person is aligned with the camera; an image modifier that modifies an image of an eye portion in the face image determined not to be a camera-eye-aligned image by the determiner to an image where the eye direction is aligned with the camera; and a controller that allows the storage device to store the shot image modified by the image modifier.

Abstract: A vehicle-mounted display system for enhancing a driver's forward viewability, which comprises a forwardly directed camera mounted within an interior of the vehicle, for imaging a forwardly directed expected field of view (EFOV) of the driver; an image generator; and a processing unit in data communication with both the forwardly directed camera and the image generator. The processing unit is operable to monitor a parameter indicative of a driver's forward viewability through a front windshield of a road over which the vehicle advances during a transportation operation and to command the image generator to generate when a value of the monitored parameter deviates from a predetermined parameter range, within the vehicle interior, an EFOV-related image which is visible to the driver and is based on image data received from the forwardly directed camera, to ensure the driver's forward viewability during the course of a transportation operation.

Abstract: Techniques disclosed herein relate to a near-eye display system. One example of an optical device of a near-eye display includes a substrate and holographic grating conformally coupled to a surface of the substrate. The substrate is transparent to visible light and infrared light and is configured to be placed in front of an eye of a user of the near-eye display. A refractive index modulation of the holographic grating is apodized in a surface-normal direction of the substrate to reduce optical artifacts in the visible light.

Abstract: Systems and methods are disclosed for operating a head-mounted display system based on user perceptibility. The display system may be an augmented reality display system configured to provide virtual content on a plurality of depth planes by presenting the content with different amounts of wavefront divergence. Some embodiments include obtaining an image captured by an imaging device of the display system. Whether a threshold measure or more of motion blur is determined to be exhibited in one or more regions of the image. Based on a determination that the threshold measure or more of motion blur is exhibited in one or more regions of the image, one or more operating parameters of the wearable display are adjusted. Example operating parameter adjustments comprise adjusting the depth plane on which content is presented (e.g., by switching from a first depth plane to a second depth plane), adjusting a rendering quality, and adjusting power characteristics of the system.

Abstract: A neural network apparatus includes processing circuitry and memory. The memory stores a plurality of images of a target. The processing circuitry is configured to: access, from the memory, a first image and an identification of a centroid pixel of the target within the first image; generate, based on a geometry of the target and the centroid pixel, a confidence map indicating, for each pixel in the first image, a confidence value that the pixel includes the target; train, using the plurality of images of the target, including the first image and the confidence map, an artificial neural network to identify the target in visual data; and provide an output representing the trained artificial neural network.

Abstract: The present disclosure describes systems and methods for acquiring a biometric image. A biometric camera system can position a lens of a biometric camera between a pixel array of the biometric camera and an iris. The pixel array can acquire an image of the iris using light reflected from the iris and transmitted through the lens of the biometric camera. The lens can increase a pixels per iris (PPi) value of the image of the iris acquired by the pixel array, by applying optical positive distortion to the light reflected from the iris when the light is optically directed through the lens. A processor can provide a biometric image for biometric matching, by image-processing the acquired image of the iris having the increased PPi value, with an inverse function of the optical positive distortion.

Abstract: Systems and methods pertaining to iris recognition are disclosed. A system is disclosed comprising an iris imager configured to acquire an iris image, and a processor configured to receive a first plurality of iris images from the iris imager, generate a first plurality of iris codes corresponding to the first plurality of iris images, generate a distribution metric corresponding to an iris cell location, generate a first composite iris code using the distribution metric, and generate a first match value using the first composite iris code and a first stored iris code.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for authenticating a user based on passive affective and knowledge-based authentication (AKBA). In one aspect, a method includes data associated with eye movements and ocular dynamics of the user are captured with a camera as the user looks at a graphical user interface (GUI) of a device; an AKBA signature of the user is determined based on the captured data; the user is authenticated based on a comparison of the AKBA signature with an AKBA template associated with the user; and an access to a subset of functions of an application is granted.

Abstract: An eyetracker obtains a digital image representing at least one eye of a subject. The eyetracker then searches for pupil candidates in the digital image according to a search algorithm and, based on the searching, determines a position for the at least one eye in the digital image. The eyetracker also obtains light-intensity information expressing an estimated amount of light energy exposing the at least one eye when registering the digital image. In response to the light-intensity information, the eye-tracker determines a range of pupil sizes. The search algorithm applies the range of pupil sizes in such a manner that a detected pupil candidate must have size within the range of pupil sizes to be accepted by the search algorithm as a valid pupil of the subject.

Abstract: Embodiments of the present disclosure disclose a method, electronic device, and computer readable medium for image identification. The method comprises: acquiring an image comprising a person object for use as an input image; performing feature extraction on the input image using a feature extracting module of a trained human body identification model; and matching an extracted human body feature of the inputted image with a preset human body feature database, to identify the person object in the inputted image, wherein the human body identification model extracting features of human body images captured by cameras of different categories respectively using the feature extracting module, and the human body identification model identifying whether the human body images captured by the cameras of different categories are human body images of a given person based on the extracted features of the human body images. This method improves the accuracy of multi-camera human body re-identification.

Abstract: A processor-implemented method of performing authentication includes obtaining a first biometric information of a user according to a first modality; calculating a first score based on the first biometric information; filtering the first score; determining whether the filtered first score satisfies a second condition; and selectively, based on a result of the determining, authenticating the user based on the first score and a first condition corresponding to the first modality. The second condition is different from the first condition.

Abstract: The invention provides an authentication method and system. It is particularly suited for verifying the identity of an individual prior to permitting access to a controlled resource. This may or may not be a financial resource. The invention uses biometric data relating to a user to encode and decode an identifier associated with a user. Thus the user's biometric data becomes the key for encoding and subsequently decoding the identifier. In one embodiment, the biometric data is used to generate a keypad configuration. The keypad configuration specifies the order and/or position of a plurality of keypad keys. An operable keypad and/or image of a keypad is then generated using the configuration. Thus, the individual's biometric data can be used to generate a customised keypad and/or image which can then be used to encode or decode the identifier associated with the user. A keypad or image generated from the biometric data can be used to generate a mapping between different keypad configurations.

Abstract: Methods, systems, and devices are disclosed for implementing a low-cost wearable multi-modal bio-sensing system capable of recording bio-markers and eye-gaze overlaid on world view in real-world settings. In an exemplary embodiment, a bio-sensing system uses at least two cameras and one or more bio-sensors to record a variety of events and bio-marker data. In another exemplary embodiment, the recorded information is used to track the eye position, calculate the pupil dimensions, and calculate the gaze of a human being. The eye position and dimensions of the pupil can be used for emotion recognition.

Abstract: The technology described in this document can be embodied in a method for preventing access to a secure system based on determining a captured image to be of an alternative representation of a live person. The method includes capturing an image of a subject illuminated by an infrared (IR) illumination source, and extracting, from the image, a portion representative of an iris of the subject. The method also includes determining that an amount of high-frequency features in the portion of the image satisfies a threshold condition indicative of the image being of an alternative representation of a live person, and in response, identifying the subject in the image to be an alternative representation of a live person. Responsive to identifying the subject in the image to be an alternative representation of a live person, the method further includes preventing access to the secure system.

Abstract: An information processing method includes the following performed using a computer: acquiring images created through capturing at the same time, and positions and capturing directions of image capturing devices that created respectively each of the images; acquiring object detection results obtained respectively through object detection processes performed using each of the images; and performing a training data selection process using the images, the positions and the capturing directions, and the object detection results acquired. The training data selection process includes: calculating a common region that is seen in the images in common, based on the positions and the capturing directions; determining a degree of agreement among the object detection results in the common region; and selecting an image to be used as training data from among the images, according to the degree of agreement.

Abstract: An example embodiment includes: an extraction unit that extracts a determination object image including a side part of an outer circumference of an iris from an image including an eye; and a determination unit that determines whether or not a colored contact lens is worn based on the determination object image.

Abstract: The technology described in this document can be embodied in a method for generating an iris identifier. The method includes obtaining a plurality of images of an iris, and generating a binary code for each of the plurality of images of the iris, the binary code including a sequence of bits. The method also includes identifying a first pattern of bits for which bit values and bit-locations are the same across a plurality of the binary codes, generating a first index based on the first pattern of bits, and then storing the first index on a storage device in accordance with a database management system. The first index is linked to biometric information of a different modality for a corresponding user.

Abstract: A method and an apparatus for recognizing an iris is provided. The method includes acquiring an image from an iris recognition camera in response to receiving a request for iris recognition, displaying a preview image including an iris recognition area corresponding to the iris recognition camera acquired from a regular camera in response to detecting an iris in the acquired image and obtaining information for the detected iris from the iris recognition camera and performing iris recognition using the information for the detected iris.

Abstract: A method of processing which can include receiving input signals from a plurality of sources, data mining the input signals, interpreting the mined data, optionally providing a set of settings and processing the input signals based on the interpretation of the mined data and, optionally, the set of settings to produce output signals. The input signals can be data mined to identify one or both of unique characteristics and attributes associable with the input signals. The mined data can be interpreted in a manner so as to form association based on one or both of the unique characteristics and attributes. The output signals can correspond to adaptably altered input signals which can be associated with adaptable alteration of one or both of visual perception and audio perception of the input signals.

Abstract: An equipment control device includes a person region detector that receives a thermal image corresponding to an object of image capture and detects a person region that is a region occupied by a person in the thermal image, a person feature detector that calculates basal metabolism of the person, based on a thermal image part of the person region in the thermal image, and an equipment controller that controls equipment, based on the basal metabolism of the person detected by the person feature detector.

Abstract: An iris detection device includes a candidate detection unit, a candidate narrowing-down unit, and an iris specifying unit. The candidate detection unit detects circular regions as iris candidates from images captured by an infrared camera. The candidate narrowing-down unit narrows down the detected iris candidates into a predetermined number of iris candidates with use of data items of images of surroundings of each of the iris candidates. The iris specifying unit specifies one of the predetermined number of iris candidates obtained by the narrowing-down as an iris by matching between gradients of a contour of each of the predetermined number of iris candidates and iris templates.

Abstract: A computer-implemented method is provided for image-based, self-guided object detection. The method includes receiving, by a processor device, a set of images. Each of the images has a respective grid thereon that is labeled regarding a respective object to be detected using grid level label data. The method further includes training, by the processor device, a grid-based object detector using the grid level label data. The method also includes determining, by the processor device, a respective bounding box for the respective object in each of the images, by applying local segmentation to each of the images. The method additionally includes training, by the processor device, a Region-based Convolutional Neural Network (RCNN) for joint object localization and object classification using the respective bounding box for the respective object in each of the images as an input to the RCNN.

Abstract: Authentication methods, systems and computer readable storage medium are provided. In an embodiment, an authentication method includes obtaining from an authentication template a first template portion, including an encoding of features, and a second template portion, including an identification of a location in the first template portion that is not usable. The method further includes revising the first template portion by setting a value at the location to a selected value to form a revised first template portion. Also, the method includes creating an authentication codeword from the revised first template portion and from public recovery data. The method further includes performing an error correction process to generate a corrected authentication codeword from the authentication codeword and from the second template portion and decoding the corrected authentication codeword.

Abstract: This disclosure describes, in part, systems and techniques for performing eye tracking. For instance, a system may include a first imaging device that generates first image data. The system may then analyze the first image data to determine a location of a face of a user. Using the location, the system may cause an actuator to move from a first position to a second position in order to direct a second imaging device towards the face of the user. While in the second position, the second imaging device may generate second image data representing at least the face of the user. The system may then analyze the second image data to determine a gaze direction of the user. In some instances, the first imaging device may include a first field of view (FOV) that is greater than a second FOV of the second imaging device.

Abstract: A system performs identification system enrollment and validation and/or authentication. In some examples, the system receives a digital representation of a biometric for a person from a device, biographic information for the person, and monitoring of capture of the biometric and determines whether the biometric is genuine. In others, the system obtains data from a network search using the biometric and the information, generates a comparison of the biometric and the information to the data from the network search, and uses the comparison to determine whether the person is who the person asserts. In still others, the system performs a number of identity checks using the information and the biometric, weights certainty and risk of each of the number of identity checks, and determines whether to approve the person for identification system enrollment using an aggregation of the identity checks.

Abstract: The present disclosure describes systems and methods for iris liveness determination. A sensor can acquire a first image of an iris with pupillary constriction, and a second image of the iris without the pupillary constriction. A processor can extract an iris portion of the first image into a first iris image, and an iris portion of the second image into a second iris image. The processor can normalize the first iris image and the second iris image. The normalized first iris image can have same dimensions as that of the normalized second iris image, in pixels. The processor can compute a minimum average difference between values of specific pixels of the normalized first iris image and values of corresponding pixels of the normalized second iris image. The specific pixels can correspond to an innermost annular portion of the iris.

Abstract: Embodiments facilitate prediction of anti-vascular endothelial growth (anti-VEGF) therapy response in DME patients. A first set of embodiments discussed herein relates to training of a machine learning classifier to determine a prediction for response to anti-VEGF therapy based on a set of graph-network features and a set of morphological features generated based on FA images of tissue demonstrating DME. A second set of embodiments discussed herein relates to determination of a prediction of response to anti-VEGF therapy for a DME patient (e.g., non-rebounder vs. rebounder, response vs. non-response) based on a set of graph-network features and a set of morphological features generated based on FA imagery of the patient.

Abstract: Exemplary embodiments are directed to biometric analysis systems generally including one or more illumination sources, a camera, and an analysis module. The illumination sources are configured to illuminate at least a portion of a face of a subject. The camera is configured to capture one or more images of the subject during illumination of the face of the subject. The analysis module is configured to analyze the one or more images captured by the camera to determine an indication of liveliness of the subject and prevent spoofing.

Abstract: The present disclosure relates to a method and apparatus for image processing, and a computer-readable storage medium. The method includes the following actions. Face recognition is performed on an image to be processed to recognize a face region in the image to be processed. Facial feature recognition points in the face region are acquired. Coordinate values of the facial feature recognition points in a coordinate system are acquired. Distances between the facial feature recognition points and angles of a coordinate axis of the coordinate system relative to lines each connecting two of the facial feature recognition points are determined according to the coordinate values. A deviation angle of the face region is determined according to the distances between the facial feature recognition points and the angles of the coordinate axis relative to the lines each connecting two of the facial feature recognition points. A retouching template corresponding to the deviation angle is acquired.

Abstract: A method includes obtaining, by a head mounted display, a trigger instruction and determining coordinates of a line of sight focus of a user on a screen of a portable electronic device, where the trigger instruction is used to instruct the head mounted display to perform a display operation, sending the coordinates to the portable electronic device, when the coordinates trigger the portable electronic device to determine that an object corresponding to the coordinates can be processed by the portable electronic device, obtaining, by the portable electronic device, related content of the object, obtaining the related content of the object of the portable electronic device, and displaying the related content of the object on a virtual screen.

Abstract: Provided is a light source device including a substrate, a first light emitting element disposed on the substrate and including a first reflective layer, a second light emitting layer configured to emit light of a second wavelength, a first etch stop layer, a first light emitting layer configured to emit light of a first wavelength different from the second wavelength, and a first nanostructure reflective layer, and a second light emitting element disposed on the substrate, spaced apart from the first light emitting element, and comprising a second reflective layer having same material and thickness as the first reflective layer, a third light emitting layer having same material and structure as the second emitting layer and configured to generate light of the second wavelength, a second etch stop layer having same material and thickness as the first etch stop layer, and a second nanostructure reflective layer.

Abstract: A blockchain may include various transactions which are identified and which require processing. The order of processing such transactions may be optimized by examining content of the transactions. One example operation may include one or more of identifying blockchain transactions, designating each of the blockchain transactions as an independent transaction type or a dependent transaction type, and determining an order to process the blockchain transactions based on the independent transaction type or the dependent transaction type.

Abstract: An image acquisition system determines first and second sets of points defining an iris-pupil boundary and an iris-sclera boundary in an acquired image; determines respective ellipses fitting the first and second sets of points; determines a transformation to transform one of the ellipses into a circle on a corresponding plane; using the determined transformation, transforms the selected ellipse into a circle on the plane; using the determined transformation, transforms the other ellipse into a transformed ellipse on the plane; determines a plurality of ellipses on the plane for defining an iris grid, by interpolating a plurality of ellipses between the circle and the transformed ellipse; moves the determined grid ellipses onto the iris in the image using an inverse transformation of the determined transformation; and extracts an iris texture by unwrapping the iris and interpolating image pixel values at each grid point defined along each of the grid ellipses.

Abstract: A computer-implemented certificate verification method includes: obtaining, by a certificate verification module, at least two images of a certificate, in which the at least two images are acquired under different acquisition conditions; obtaining, from the at least two images, at least two target images that correspond to respective images of the at least two images and that each include an image of a light-reflective coating of the certificate; and determining, based on the at least two target images, a probability that the certificate is an original.

Abstract: A system for vision in virtual reality or augmented reality is provided, in a support to be mounted on a head of a user, the system including a matrix of light-emission elements extending according to a display surface for display of images configured to be seen by the user; an optical projection system to form an image of the elements on a focusing surface at a distance from an eye of the user; a matrix of photo-detectors to acquire an image of a region of the eye extending according to a capture surface and arranged coplanar with the elements and at least partially overlapping with the capture surface and the display surface; and a computer to receive as an input the image acquired, and to process the same to deduce a parameter relative to the user, the optical projection system being a variable-focal-length optical system.

Abstract: Described is neuromorphic system for authorized user detection. The system includes a client device comprising a plurality of sensor types providing streaming sensor data and one or more processors. The one or more processors include an input processing component and an output processing component. A neuromorphic electronic component is embedded in or on the client device for continuously monitoring the streaming sensor data and generating out-spikes based on the streaming sensor data. Further, the output processing component classifies the streaming sensor data based on the out-spikes to detect an anomalous signal and classify the anomalous signal.

Abstract: Occlusion of facial features may be detected and assessed in an image captured by a camera on a device. Landmark heat maps may be used to estimate the location of landmarks such as the eyes, mouth, and nose of a user's face in the captured image. An occlusion heat map may also be generated for the captured image. The occlusion heat map may include values representing the amount of occlusion in regions of the face. The estimated locations of the eyes, mouth, and nose may be used in combination with the occlusion heat map to assess occlusion scores for the landmarks. The occlusion scores for the landmarks may be used control one or more operations of the device.

Abstract: There is provided an information processing apparatus to make it possible to dynamically change the blinking pattern of the light source according to the imaging result, the information processing apparatus including: a light source control unit configured to control each of light sources constituting a plurality of light sources that irradiate an eye with light; and an image acquisition unit configured to acquire a captured image of the eye including a bright spot that is a reflection point of the light. The light source control unit controls blinking of each of the light sources on a basis of the captured image of the eye.

Abstract: An image processing method for iris recognition of a predetermined subject, comprises acquiring through an image sensor, a probe image illuminated by an infra-red (IR) illumination source, wherein the probe image comprises one or more eye regions and is overexposed until skin portions of the image are saturated. One or more iris regions are identified within the one or more eye regions of said probe image; and the identified iris regions are analysed to detect whether they belong to the predetermined subject.

Abstract: The invention provides an identity recognition method and device. The method includes that: an eye image when a user to be recognized looks at a target point position is acquired; identity recognition is performed on the user to be recognized according to the eye image and a preestablished user identity information database; and living body detection is performed on the user to be recognized according to the eye image, a coordinate of the target point position, the user identity information database and an identity recognition result.

Abstract: In an embodiment, a method comprises determining, by a portable device, image capturing conditions by analyzing contents of a particular digital image of one or more first digital images captured by one or more image capturing devices. It is also determined whether the image capturing conditions indicate indoor image capturing conditions. If the image capturing conditions indicate indoor image capturing conditions, then it is determined whether the particular digital image includes a depiction of at least one eye. If so, an iris region, depicting the at least one eye, is segmented in the particular digital image. If the segmented iris region does not include iris valid information, then one or more image capturing devices capture one or more second digital images having an enhanced contrast. If the one or more second digital images include iris valid information, then iris code is extracted from the one or more second digital images.

Abstract: A biometric authentication device includes: an acquisition device configured to acquire eye information of a user; a memory; and a processor coupled to the memory and the processor configured to: extract a first eye feature based on recognition characteristic from the eye information; extract a second eye feature based on eye movement characteristic from the eye information; and match a first eye feature enrolled in advance with the first eye feature extracted by the first extractor, and to match a second eye feature enrolled in advance with the second eye feature extracted by the second extractor.

Abstract: A method generates depth maps at a camera having illuminators, a lens assembly, an image sensing element, a processor, and memory. The illuminators operate in a first mode to provide illumination, the lens assembly focuses incident light on the image sensing element, the memory stores programs for execution by the processor, and the processor executes the programs to control operation of the camera. The method reconfigures the illuminators to operate in a second mode, where each of a plurality of subsets of the illuminators provides illumination of a scene separately. For each subset, the process activates the illuminators in the subset without activating illuminators not in the subset and receives reflected illumination from the scene incident on the lens assembly and focused onto the image sensing element. The measured light intensity values of the received reflected illumination at the image sensing element are transmitted to a remote server for processing.