Abstract: A method for recognizing a face from monitoring video data is disclosed. Two neural networks are used to compare and score high-dimensional face features of a face, and a K-neighbor algorithm and a screening mechanism with a Euclidean distance as a threshold are combined for face comparison recognition to obtain an accurate face recognition result. In addition, the present disclosure also performs further screening based on the time of video data and the frequency of face appearance, and finally obtains a recognition result, thereby obtaining a more accurate face recognition result. The present disclosure can perform relatively accurate face recognition on video data or picture data captured by a real-time or historical monitoring camera.

Abstract: An electronic device and method are disclosed. The electronic device includes a cover glass, a display panel, a fingerprint sensor, and a processor. The processor implements the method, including: obtaining, through a fingerprint sensor, a fingerprint image from an external object contacting a surface of the electronic device, detecting, using a processor, a feature point of the fingerprint image, comparing the detected feature point of the fingerprint image with a feature point of a pre-stored reference image, detecting a variation in a distance between the surface of the electronic device and the fingerprint sensor based on a result of the comparison, and recognizing the fingerprint based on the detected variation in the distance.

Abstract: In a method for operating a fingerprint sensor including a plurality of ultrasonic transducers and a presence sensor, the presence sensor proximate the fingerprint sensor is activated, where the presence sensor is for detecting interaction between an object and the fingerprint sensor. Subsequent to detecting interaction between an object and the fingerprint sensor at the presence sensor, a subset of ultrasonic transducers of the fingerprint sensor is activated, the subset of ultrasonic transducers for determining whether the object is a human finger, where the subset of ultrasonic transducers is proximate the presence sensor such that the subset of ultrasonic transducers and the presence sensor can concurrently interact with the object.

Abstract: The disclosure relates to systems, methods and processor-readable storage mediums, having processor-executable instructions stored thereon, for determining the position of a real-world object in an augmented reality application running on a computing device having a camera. The method includes capturing image information using the camera. Detection of real-world objects is then performed using a computer vision model, and filtering of a received point cloud is performed by way of a virtual frustum constructed using the captured image information. Finally, the position of the object is determined using the filtered points.

Abstract: A method of generating a finger image includes determining a quantity of electric charge to be removed from each of a plurality of detection circuits connected to a fingerprint sensor, based on an amplifier characteristic of each of the plurality of detection circuits; obtaining a second electrical quantity by removing the quantity of electric charge from a first electrical quantity that is input to each of the plurality of detection circuits; integrating the second electrical quantity to obtain an integrated value; and generating the fingerprint image based on comparison between the integrated value of the second electrical quantity and a predetermined threshold value.

Abstract: The present disclosure relates to methods and apparatuses for liveness detection, electronic devices, and computer readable storage media, improving the accuracy of liveness detection. The method includes: carrying out feature extraction on a collected image to obtain image feature information; determining a modality classification result of the image based on the image feature information, the modality classification result indicating that the image corresponds to a target modality in at least one modality; and determining whether a target object in the image is living based on the modality classification result of the image.

Abstract: The present invention generally relates to a method for authenticating a user of an electronic device comprising a fingerprint sensor. The method comprises acquiring a sequence of verification images. Estimating, based on the images a first and a second verification force value indicative of a force applied by the object on the fingerprint sensor when the verification images were acquired. A verification image representation and a verification force value forms a verification template for each of the first and second images and forces. Matching the first and second verification template with at least one of a plurality of stored fingerprint enrolment templates in a respective first and second matching step. When the first and second matching steps each results in a positive match between the respective verification template and a respective one of the plurality of stored fingerprint enrolment templates, providing a signal indicating successful authentication of the user.

Abstract: The present invention discloses a multi-camera multi-face video splicing acquisition device, comprising a plurality of cameras for successively capturing video or images with faces, at least one splicing server for face tracking, face recognition, face cutting, face sorting and face splicing of face videos or images captured by cameras; and at least one time synchronizer for calibrating the time of at least one camera and splicing server; the above devices are interconnected through a network to achieve data interaction with each other. By serially splicing the face images of the same person acquired by multiple cameras, a face sequence of a longer period of time can be formed, and the face sequences sorted by time series could be used to further extract feature information for various time series analysis, and the longer the length of time of the face sequence, the more valid information can be extracted after time-series analysis.

Abstract: Methods and systems for depth sensing are provided. A system includes a first and second optical sensor each including a first plurality of photodetectors configured to capture visible light interspersed with a second plurality of photodetectors configured to capture infrared light within a particular infrared band. The system also includes a computing device configured to (i) identify first corresponding features of the environment between a first visible light image captured by the first optical sensor and a second visible light image captured by the second optical sensor; (ii) identify second corresponding features of the environment between a first infrared light image captured by the first optical sensor and a second infrared light image captured by the second optical sensor; and (iii) determine a depth estimate for at least one surface in the environment based on the first corresponding features and the second corresponding features.

Abstract: A fingerprint recognition card is provided to perform authentication and security functions by recognizing a fingerprint of a user and includes a fingerprint recognition unit configured to detect a fingerprint of a user, a control unit configured to perform registration authentication for the detected fingerprint of the user, a communication unit configured to perform tagging to an external reader depending on the registration authentication of the control unit, and an inductive current generator configured to generate an inductive current in response to approaching the external reader. The inductive current generator generates the inductive current when a distance to the external reader is within a predetermined distance, converts the generated inductive current to a direct current (DC), and supplies the output voltage, which is generated by reducing the input voltage of the converted DC, to the fingerprint recognition unit, the control unit, and the communication unit.

Abstract: A system and methodology provide for annotating videos with entities and associated probabilities of existence of the entities within video frames. A computer-implemented method identifies an entity from a plurality of entities identifying characteristics of video items. The computer-implemented method selects a set of features correlated with the entity based on a value of a feature of a plurality of features, determines a classifier for the entity using the set of features, and determines an aggregation calibration function for the entity based on the set of features. The computer-implemented method selects a video frame from a video item, where the video frame having associated features, and determines a probability of existence of the entity based on the associated features using the classifier and the aggregation calibration function.

Abstract: Provided is a process of classifying ambient signals to reduce fraudulent use of information, the process including: receiving, with the mobile computing device, from a remote server, balance-access information by which a stored value card balance can be spent at a point of sale terminal; storing, with the mobile computing device, the balance-access information; sensing, with one or more sensors of a mobile computing device, ambient signals; classifying the ambient signals as indicating the user is in a retail establishment; and in response to the classification, displaying, on a display screen of the mobile computing device, the balance-access information, such that the balance-access information can be input to a point-of-sale terminal.

Abstract: An information processing system includes: a recognition unit configured to perform recognition processing on a video captured by an image capturing unit; a determination unit configured to determine timing to update a model used for the recognition processing; and an update unit configured to update the model at the determined timing.

Abstract: An image processing apparatus sets, for at least one portion of a first image in which a brightness range is a first range, relationship information which associates an input brightness value with an output brightness value, changes, based on the relationship information, brightness values of pixels in the portion, and outputs a second image which has a second range of brightness. The relationship information is generated so as to associate, with each of the brightness values of the pixels in the portion, an output brightness value which does not exceed a value higher than the brightness value by a predetermined value, and to associate, with a brightness value which is not less than a feature amount among the brightness values of the pixels in the portion, a predetermined output brightness value falling within a predetermined range.

Abstract: Described herein are systems and techniques for generating training data for use in training a machine learning model for image enhancement. The system may access a target image of a displayed video frame, wherein the target image represents a target output of the machine learning model. The system may access an input image of the displayed video frame, wherein the input image corresponds to the target image and represents an input to the machine learning model. The system may train the machine learning model using the target image and the input image corresponding to the target image to obtain a trained machine learning model.

Abstract: A face liveness recognition method includes: obtaining a target image containing a facial image; extracting facial feature data of the facial image in the target image; performing face liveness recognition according to the facial feature data to obtain a first confidence level using a first recognition model, the first confidence level denoting a first probability of recognizing a live face; extracting background feature data from an extended facial image, the extended facial image being obtained by extending a region that covers the facial image; performing face liveness recognition according to the background feature data to obtain a second confidence level using a second recognition model, the second confidence level denoting a second probability of recognizing a live face; and according to the first confidence level and the second confidence level, obtaining a recognition result indicating that the target image is a live facial image.

Abstract: An online community offering communication, guidance, and features used to enhance photographs to reveal eyes, faces, images, religious iconery, spirits, animals and more is presented herein. Each of the revealed images is already embedded in every photograph. The present invention allows users to reveal those images. The processes for revealing hidden images involve removing different layers of the photograph to enhance what is already there and hidden, hence revealing the hidden spiritual images.

Abstract: A method for fingerprint enrollment, a terminal, and a non-transitory computer readable storage medium are provided. The method includes the following. An enrolled fingerprint image is acquired. The enrolled fingerprint image is compared with a preset fingerprint template. Fingerprint ID of the preset fingerprint template is determined as fingerprint ID of the enrolled fingerprint image when the enrolled fingerprint image matches the preset fingerprint template. New fingerprint ID is generated as the fingerprint ID of the enrolled fingerprint image when the enrolled fingerprint image fails to match the preset fingerprint template.

Abstract: A method of entering a facial image of an individual into a database of recognised and authorised individuals, the method comprising the steps of: receiving, from an image capture device connected to a database processing device, an image of a scene; determining whether the image of the scene contains a facial image of a recognised and authorised first individual; determining whether the image of the scene contains a facial image of a second individual and, if the image of the scene contains a facial image of a second individual, determine if the second individual is a recognised and authorised individual; if the second individual is not a recognised and authorised individual, determine whether the image of the scene includes a security gesture by the recognised and authorised first individual; and entering the facial image of the second individual to the database of recognised and authorised individuals if the image of the scene includes the security gesture by the recognised and authorised first individual.

Abstract: There is provided a survey marker, and an image processing apparatus, method, and program capable of accurately detecting a survey marker from a captured image obtained by image capturing. The survey marker has a planar shape and includes a plurality of circles concentrically disposed, the plurality of circles including adjacent circles each having a different color. A candidate region extraction unit extracts a candidate region from a captured image obtained by image capturing of the survey marker, the candidate region being a candidate of a region in which the survey marker appears. A feature amount extraction unit extracts a feature amount of the candidate region. A discrimination unit discriminates the survey marker on the basis of the feature amount. The present technology can be applied to, for example, a case of detecting a survey marker installed on the ground from a captured image obtained by aerial image capturing.