Abstract: Systems and methods may generally be used to authenticate a user. An example method may include capturing an image of a portion of a face of a user, retrieving, based on identification information, a registered model of the face from a database, and comparing the portion of the face in the captured image to a facial feature of the registered model. In response to determining that the portion of the captured image matches the facial feature of the registered model, the method may include granting access to secure content.

Abstract: An electronic device is disclosed that includes a camera, a display, and a processor.

Abstract: Embodiments of the present disclosure relate to the field of image processing technologies and disclose an image processing method and apparatus, an electronic device, and a computer-readable storage medium. The image processing method includes: when an obtained first image includes a human face, performing a first transformation process on the first image to obtain a second image; determining, based on a first target face key point of the human face in the first image, a target position, in the first image, of a second target face key point of the human face in the second image; performing a first movement process on the second image based on the target position; and generating a target image based on the first image and the second image processed through the first movement process, and displaying the target image.

Abstract: Systems and methods for distorting CAPTCHA images with generative adversarial networks include an image distortion interface that can select an image record from a database, determine the size of the selected image, and apply an adversarial attack algorithm to create an array of pixels that is the same size as the selected image so that the majority of the pixels in the array are zero values and the remaining pixels in the array are in the red-green-blue value range. The image distortion interface can merge the array of pixels with the selected image to form a distorted image. A server can include a CAPTCHA test interface that can select a set of images having the same label, retrieve corresponding distorted images, and provide a CAPTCHA test using the distorted images. The CAPTCHA test can use distorted images with different applied adversarial attack algorithms.

Abstract: This application discloses a password verification method and a password setting method. The password verification method includes: in response to a detected operation of requesting for password verification, collecting at least one first image by using a camera of a mobile terminal; obtaining matching information when a result of matching between the at least one first image collected by the mobile terminal and at least one first preset image satisfies a first preset matching condition, where the matching information includes at least one of the following: location information of the mobile terminal, motion information of the mobile terminal, at least one second image collected by the camera of the mobile terminal, and network connection information of the mobile terminal; and performing matching between the obtained matching information and a second preset matching condition, where the password verification succeeds when the matching is successful.

Abstract: A system for providing credential activation layered security is disclosed. In particular, the system adds a layer of additional security at ingress and egress points of a location, such as a building. When a user attempts to check in at the location, the user may provide a proof of physical presence, a proof of digital presence, or a combination thereof, such as at a device at the location. In order to activate a credential for accessing physical and/or logical access control systems of the location, the system may authenticate the proof of physical presence, the proof of digital presence, or both. If the system authenticates the user, the user may be checked-in and the credential may be activated so that the user may access the physical and/or logical access control systems of the location so as to gain access to the ingress point or exit via the egress point.

Abstract: A graphics processor can include a processing cluster array including a plurality of processing clusters coupled with the plurality of memory controllers, each processing cluster of the plurality of processing clusters including a plurality of streaming multiprocessors, the processing cluster array configured for partitioning into a plurality of partitions. The plurality of partitions include a first partition including a first plurality of streaming multiprocessors configured to perform operations for a first neural network, The operations for the first neural network are isolated to the first partition. The plurality of partitions also include a second partition including a second plurality of streaming multiprocessors configured to perform operations for a second neural network. The operations for the second neural network are isolated to the second partition and protected from operations performed for the first neural network.

Abstract: Methods, devices, and systems related to a user interface for modifying pictures are described. Modifying pictures can include placing an image of a picture in a different picture, replacing an image of the picture with a different image, deleting an image from the picture, and/or storing an image from the picture. In an example, a method can include displaying a first picture on a user interface of a mobile device, receiving, on the user interface, a selection of at least one foreground image included in the first picture, receiving, on the user interface, a selection of a background image included in a library of images, and displaying, on a user interface, a second picture including the at least one selected foreground image and the selected background image responsive to receiving the selection of the at least one foreground image followed by the selection of the background image.

Abstract: The present technique relates to an apparatus and a method for video-audio processing, and a program each of which enables a desired object sound to be more simply and accurately separated. A video-audio processing apparatus includes a display control portion configured to cause a video object based on a video signal to be displayed; an object selecting portion configured to select the predetermined video object from the one video object or among a plurality of the video objects; and an extraction portion configured to extract an audio signal of the video object selected by the object selecting portion as an audio object signal. The present technique can be applied to a video-audio processing apparatus.

Abstract: An information processing device according to the present invention includes a person extraction means that extracts a person in a captured image, an action extraction means that extracts an action of a person group including a plurality of persons other than a given person in the captured image, and an identification means that identifies a given person group based on a result of extraction of the action of the person group.

Abstract: The present disclosure provides a method and device to process identity information based on a fundus image. An identity information comparison method includes: recognizing a fundus image by using a neural network to obtain a multi-dimensional feature vector representing the identity of a user; comparing the obtained multi-dimensional feature vector with each multi-dimensional feature vector stored in a database; and determining, according to a comparison result, whether there is a matching between the currently obtained multi-dimensional feature vector and a multi-dimensional feature vector pre-stored in the database.

Abstract: An imaging device with a plurality of image sensing pixels and a plurality of event detection pixels is provided. The image sensing and event detection pixels can be provided as part of different arrays of pixels, or can be included within a common array of pixels. The event detection pixels are operated continuously to provide signals indicating the occurrence of events. In response to detecting an event, image sensing pixels are selectively operated. The selective operation of the image sensing pixels can include activation of image sensing pixels within one or more regions of interest, while image sensing pixels not included in any region of interest remain off. The selective operation of the image sensing pixels can include the selection of a frame rate applied across an array of mage sensing pixels, or within determined regions of interest.

Abstract: A face authentication machine includes a face image acquirer (camera) that acquires a live image of a predetermined capturing area in a facility including a face image of a user; a display unit (display) that displays a progress of face authentication of the face image included in the live image acquired by the face image acquirer; and a processor (controller) that controls the face image acquirer and the display unit, in which the processor sequentially displays two or more notification screens indicating the progress on the display unit, when the user is detected from the live image.

Abstract: A processor-implemented image processing method and apparatus are provided. The image processing method includes receiving an input image and rotation information associated with the input image, generating a feature vector of the input image based on pose information corresponding to the input image, generating an assistant feature vector which represents a target component according to a pose corresponding to the rotation information, based on the feature vector, the pose information, and the rotation information, and generating a target image which has the pose corresponding to the rotation information based on the feature vector and the assistant feature vector.

Abstract: Systems and methods are provided for enhanced animation generation based on generative modeling. An example method includes training models based on faces and information associated with persons, each face being defined based on location information associated with facial features, and identity information for each person. The modeling system being trained to reconstruct expressions, textures, and models of persons.

Abstract: A process and an activation device (30) provides a user (B) activation of a drug measuring device (100). This drug measuring device includes an input unit (3), a testing unit (6) for testing given samples and an image recording device (12). The activation device (30) receives an activation image set (InB), which was generated by the drug measuring device or by a drug measuring device and at least one image (40, 40.1) from the image recording device. The received activation image set is checked as to meeting a predefined release criterion. For this the activation image set is compared with a predefined reference image set (35) for the user. If the release criterion is met, the drug measuring device is released for the user. An activation data set (32) for the user is completed by at least one image of the activation image set, which shows the user.

Abstract: A system and method for image recognition registration of a person into an event. The person is registered in the event using image recognition technology. A digital image of the person taken by a camera is compared with a stored profile image of the person to identify the person and enter them into the event without the need for the person to pre-register for the particular event. Enhanced recognition techniques incorporating pattern recognition may be used to increase identity accuracy.

Abstract: There is provided a method of determining a three-dimensional shape of an insert for insertion into an ear. The method includes receiving image data corresponding to a two-dimensional image of an ear, processing the image data to measure at least one biometric feature of the ear, the at least one biometric feature being indicative of a three-dimensional shape of at least part of the ear, and determining a three-dimensional shape of an insert for insertion into the ear by matching said at least one biometric feature with one of a plurality of pre-stored three-dimensional shapes. Each pre-stored three-dimensional shape corresponds to a respective ear.

Abstract: Named Entity Recognition (NER) in a user search query in a real-time search engine may be achieved by training a machine learning algorithm to create a trained model. The trained model may be configured to receive a user search query as input and to recognize and output zero or more named entities in the search query. The training may include an iterative training process in which further training data is added at each iteration, in some embodiments. The training may be based on three training data sets, in some embodiments. A first training data set may be based on user search and engagement activity. A second training data set may be artificially generated based on a catalog of named entity values. A third training data set may be based on optimized and supplemented data pairs sampled from the first training data set.

Abstract: Traditionally, biometrics provides identification based on fingerprints, iris, or face to a vendor. The vendor's database can be hacked or a person's traditional biometrics can be stolen or compromised. Because an individual has only FFten fingers and two eyes, if compromised, secure identification becomes very difficult. A new approach for personal identification, leveraging “skin” biometrics, offers protection for individuals, and provides the ability to change their biometric identity by selecting a different patch of skin to use as skin identification. The biometric method and system provides biometric identification based on a small section of a person's skin. Skin biometrics is leveraged for positive identification of individuals. Skin biometrics captures a region of the subject's skin topography, then maps surface landmarks, creating a unique result that the user can provide in various authentication scenarios.

Abstract: An example system includes a landmark engine to detect a facial landmark in an image of a face. The system includes a comparison engine to determine a difference between the facial landmark in the image and a facial landmark of a neutral face. The system also includes an action engine to determine whether a facial action unit occurred based on whether the difference satisfies a condition.

Abstract: A neural network method and apparatus is provided. A processor-implemented neural network method includes determining, based on a determined number of classes of input data, a precision for a neural network layer outputting an operation result, and processing parameters of the layer according to the determined precision.

Abstract: An anti-counterfeiting face detection method, device and multi-lens camera are disclosed. The anti-counterfeiting face detection method comprises: acquiring a depth image, an infrared image and an RGB image by using a TOF camera and an RGB camera; analyzing the RGB image through a preset face detection algorithm to determine an RGB face region of a face in the RGB image and position information of the RGB face region; determining a depth face region of the face in the depth image and an infrared face region of the face in the infrared image based on the position information of the RGB face region; determining that the face passes the detection when the depth face region, the infrared face region and the RGB face region meet corresponding preset rules respectively.

Abstract: An electronic device is disclosed that includes a camera, a display, and a processor.

Abstract: Techniques are disclosed relating to biometric authentication, e.g., facial recognition. In some embodiments, a device is configured to verify that image data from a camera unit exhibits a pseudo-random sequence of image capture modes and/or a probing pattern of illumination points (e.g., from lasers in a depth capture mode) before authenticating a user based on recognizing a face in the image data. In some embodiments, a secure circuit may control verification of the sequence and/or the probing pattern. In some embodiments, the secure circuit may verify frame numbers, signatures, and/or nonce values for captured image information. In some embodiments, a device may implement one or more lockout procedures in response to biometric authentication failures. The disclosed techniques may reduce or eliminate the effectiveness of spoofing and/or replay attacks, in some embodiments.

Abstract: Provided are systems and methods for face reenactment. An example method includes receiving visual data including a visible portion of a source face, determining, based on the visible portion of the source face, a first portion of source face parameters associated with a parametric face model, where the first portion corresponds to the visible portion, predicting, based partially on the visible portion of the source face, a second portion of the source face parameters, where the second portion corresponds to the rest of the source face, receiving a target video that includes a target face, determining, based on the target video, target face parameters associated with the parametric face model and corresponding to the target face, and synthesizing, using the parametric face model, based on the source face parameters and the target face parameters, an output face that includes the source face imitating a facial expression of the target face.

Abstract: Segmentation first breaks the images into segments or regions, with the segments of the region having text or symbols. The segmented image is separately applied to two different CNN-based models. Each model produces text boxes where potential text might exist. Then, a selective NMS algorithm is applied to the output of each model to produce a final group of text regions. These text regions are analyzed and actions are taken.

Abstract: Techniques are described for generating virtual makeup products. Virtual makeup products, in this context, may include a predefined effect and a configuration for the effect, wherein the predefined effect includes one or more layered visual filters. In an embodiment, a system in accordance with the present disclosure may acquire makeup product definition data based on characteristics of a real-world makeup product from an entity associated with the real-world product such as a designer, manufacturer, or vendor. Using the makeup product definition data, the system may generate a virtual makeup product includes a configuration of a predefined effect based on the acquired product definition data. The generated virtual makeup product is then passed through a process of internal calibration, internal review, and external review to improve the accuracy of the simulation of the real-world makeup product. In some embodiments, external review is performed by an external reviewing entity (e.g.

Abstract: A data search system includes an agent server that includes a storage and stores data in the storage, and a host server that manages tag management information in which a tag as a term for searching the data and a storage as a storage location of the data are associated with the data, receives a query including a search designation tag, and acquires data associated with the search designation tag from the storage location of the data by referring to the tag management information and responds to the query. The host server continuously updates the tag management information according to content and the storage location of the data.

Abstract: The Selectable Facial Recognition Assistance Method and System (SFRA) allows law enforcement officers, military personnel, and other users the ability to rapidly and accurately identify wanted or known personnel, by combining the ability to select the type of examination, selection of optimized face matching algorithms and expert face examiners. The SFRA allows the user to select “Quick Looks” or “Long Looks” with respect to the level of detail of examination, along with race-based selected algorithms and expert face examiners. The race-based focused examination combining specialized algorithms and examiners will produce the highest confidence levels of match or no-match based on photographs taken, submitted, as well as database photographs. The SFRA will greatly assist users and persons being photographed to avoid false matches and false no-matches. This system will greatly reduce misidentification, profiling, or any other identification weaknesses that the systems and users may have.

Abstract: Techniques for evaluating an output of a machine learning model and using the evaluation to retrain the machine learning model are described. For example, a data set that is output from a layer of the machine learning model is reduced to a 2-D or 3-D representation that is suitable for viewing. A user views the reduced data set in a viewing environment such as virtual reality or augmented reality. The user makes changes using that viewing environment. The changes are then used to retrain the machine learning model.

Abstract: Example posture analysis systems and methods are described. In one implementation, a computing system identifies a deviation of a user's posture from a predetermined posture based on visual data associated with the user. The user is informed of the deviation via a graphical image displayed on a display device. The user is then provided with instructions for correcting the deviation via the graphical image displayed on the display device.

Abstract: An anomaly detection device is located between a network and a first ECU in the plurality of ECUs, and includes: a communication circuit; a processor; and a memory including a set of instructions that, when executed, causes the processor to perform operations including: receiving a message from the first ECU and transmitting the message to the network, and receiving a message from the network and transmitting the message to the first ECU, using the communication circuit; holding, in the memory, a received ID list; when an ID of the message received by the communication circuit from the network is not included in the received ID list, adding the ID to the received ID list; and when an ID of the message received by the communication circuit from the first ECU is included in the received ID list, causing the communication circuit not to transmit the message to the network.

Abstract: An information processing apparatus (100) includes a collation unit (102) that collates first feature information extracted from a person included in a first image (20) with first registered feature information stored in a storage unit (110), and a registration unit (104) that stores, in a case where the first feature information is unable to be extracted from the person or a collation result in the collation unit (102) indicates a mismatch, second feature information extracted from the person in the storage unit (110) as second registered feature information, in which the collation unit (102) collates second feature information extracted from a person included in a second image (22) with the second registered feature information stored in the storage unit (110), and thus specifies the person corresponding to the second registered feature information in the second image (22).

Abstract: Systems and methods are disclosed herein for applying behavioral-based parental controls for media assets. The disclosed techniques herein discuss determining a media asset viewed by a user and character actions of a character within the media asset. Behavioral user data is received which includes a user action. A determination is made whether the user action matches one of the character actions. If a match is found, a parental control restriction setting for the media asset is applied and this parental control restriction is stored in memory.

Abstract: In an embodiment, a method includes receiving a plurality of first images including at least a mouth-related portion of a human uttering a voice, wherein each first image has depth information; obtaining a noisy spectrogram including a first representation of the voice of the human; extracting a plurality of visual features using the first images, wherein one of the visual features is obtained using the depth information of a second image of the first images; extracting a plurality of audio features using the noisy spectrogram; determining a first spectrogram using the visual features and the audio features; subtracting the first spectrogram from the noisy spectrogram, to obtain a purified representation of the voice of the human; and outputting, by an I/O-related outputting device, a response using the purified representation of the voice of the human.

Abstract: The present application provides a method and apparatus of creating a face model, and an electronic device. The method includes: obtaining at least one key point feature of an current face image by performing key point detection on the current face image; obtaining a target bone parameter set matching the current face image according to the at least one key point feature; and creating a virtual three-dimensional face model corresponding to the current face image according to the target bone parameter set and a standard three-dimensional face model.

Abstract: A method and system of performing an authentication of a person in order to activate a smart card for access to a secured resource is disclosed. The system and method are configured to collect image data and compare the image data to facial model data stored on the smart card. The system detects whether there is a match between the image data and any facial signatures previously collected and stored in the smart card. If there is a match, the system verifies an identity of the person and can be configured to automatically activate the smart card for a limited period of time.

Abstract: Systems and methods for evaluating face recognition algorithms in view of image classification features affected by smart makeup are provided. Increasingly, face recognition technology is being used in applications beyond biometric identification for authentication/login purposes. Face recognition technology has been deployed as part of surveillance cameras, which may capture facial images that can be used as evidence of criminal conduct in a court of law. While admissibility of such evidence is subject to the rules of evidence used for any other piece of evidence, reliance on such face recognition technology poses challenges. Until these face recognition algorithms become properly trained and are deployed in a manner that does not result in false positives in the context of policing and judiciary, other solutions are needed. Smart makeup will improve the usage of this technology that has traditionally had a higher false positive identification rate for people of color and women.

Abstract: Disclosed is a system for face recognition based on dynamic updating of facial features, comprising an image acquisition unit, a face image standardization unit, a facial feature comparison unit, and a facial feature update unit. The image acquisition unit acquires an original image which is processed by the face image standardization unit, and then the facial feature comparison unit completes extraction and comparison of a facial feature vector to determine whether the original image belongs to a user ID or a stranger, or to complete entry of the facial feature vector. Each user ID corresponds to one or more facial feature vectors. The facial feature update unit automatically updates the facial feature vector in a normal workflow to improve reliability and accuracy of face recognition. Also disclosed is a method for face recognition using the system.

Abstract: Systems and methods for identifying objects of interest in video streams associated with an incident. One system includes an electronic processor that receives a search attribute for an object associated with the incident and receives, from a primary video analytics system having a first video coverage capability associated with the incident, a video analytics policy for the first capability. The processor identifies a conflict between the attribute and the policy, and, in response, determines a plurality of secondary video analytics systems, each having a secondary video coverage capability. The processor, for each of the plurality of secondary systems: receives a secondary video analytics policy and determines whether a secondary conflict exists between the attribute and the policy, and, in response to determining that a conflict does not exist, transmits, to the secondary video analytics system, a request to apply the search attribute to a video stream of the secondary capability.

Abstract: Techniques for providing location identifiers of devices to render image data comprising faces are described. According to an example of the present subject matter, a face present in image data is detected. A contact identifier corresponding to the face is obtained and a location identifier of a device to render image data is sent to the contact identifier.

Abstract: Techniques are described for passive three-dimensional (3D) object authentication based on image sizing, such as for biometric facial recognition. For example, during a registration routine, an imaging system captures images of a registering user's face at multiple distances. The images can be processed to extract registration dimensions, including individual deterministic structural dimensions, dimensional relationships that are static over changes in imaging distance, and dimensional relationships that changes predictably over changes in imaging distance. During an authentication routine, the imaging system again captures authentication images of an authenticating user's face (purportedly the previously registered user) at some authentication imaging distance and processes the images to extract authentication dimensions.

Abstract: Systems and methods use machine learning models with content editing tools to prevent or mitigate inadvertent disclosure and dissemination of sensitive data. Entities associated with private information are identified by applying a trained machine learning model to a set of unstructured text data received via an input field of an interface. A privacy score is computed for the text data by identifying connections between the entities, the connections between the entities contributing to the privacy score according to a cumulative privacy risk, the privacy score indicating potential exposure of the private information. The interface is updated to include an indicator distinguishing a target portion of the set of unstructured text data within the input field from other portions of the set of unstructured text data within the input field, wherein a modification to the target portion changes the potential exposure of the private information indicated by the privacy score.

Abstract: Method, an electronic device and a storage medium for living body detection based on face recognition are disclosed. The method comprises: obtaining to-be-detected infrared image and visible light image; performing edge detection and texture feature extraction on the infrared image, and feature extraction on the visible light image through a convolutional neural network; and determining whether the infrared and visible light images pass living body detection based on results of the edge detection and texture feature extraction on the to-be-detected infrared image, and a result of feature extraction on the to-be-detected visible light image through the convolutional neural network. The method, an electronic device and a storage medium for living body detection based on face recognition combine the advantages of three technologies of edge detection, texture feature extraction and convolution neural network, effectively perform living body detection, and improve the determination accuracy.

Abstract: Techniques are described for authenticating an individual based at least partly on a tremor signature of the individual. Motion data is collected through motion sensor(s), such as accelerometers, gyroscopic motion sensors, and so forth, of a portable computing device that is being held, worn, or is otherwise in contact with the user. Based on the collected motion data, a tremor signature may be determined and compared to a previously determined model for the user, and an authentication determination may be made based on the comparison. The tremor signature may be used in combination with other information to authenticate the user. For example, the tremor signature may be used in combination with location information and/or other biometric data such as a fingerprint scan, image of the user's face, audio recording of the user's voice, the user's heartbeat, a cardiac electrical signature, bio-electrical impedance, and so forth.

Abstract: Techniques related to implementing convolutional neural networks for face or other object recognition are discussed. Such techniques may include applying, in turn, a depth-wise separable convolution, a condense point-wise convolution, and an expansion point-wise convolution to input feature maps to generate output feature maps such that the output from the expansion point-wise convolution has more channels than the output from the condense point-wise convolution.

Abstract: Gesture-responsive modification of a generated sound field is described.

Abstract: An electronic device and a method for controlling thereof are provided. A method for controlling an electronic device according to the disclosure includes obtaining a plurality of images for performing clustering, obtaining a plurality of target areas corresponding to each of the plurality of images, obtaining a plurality of feature vectors corresponding to the plurality of target areas, obtaining a plurality of central nodes corresponding to the plurality of feature vectors, obtaining neighbor nodes associated with each of the plurality of central nodes, obtaining a subgraph based on the plurality of central nodes and the neighbor nodes, identifying the connection probabilities between the plurality of central nodes of the subgraph and the neighbor nodes of each of the plurality of central nodes based on a graph convolutional network, and clustering the plurality of target areas based on the identified connection probabilities.

Abstract: In a data preparation process, raw data in a database is classified into first-type data and second-type data. In a data retrieval process, a first retrieval range is determined in the first-type data, and to-be-retrieved data is retrieved in the first retrieval range, to obtain a first retrieval result; the to-be-retrieved data is retrieved from all of the second-type data, to obtain a second retrieval result; and finally, a final retrieval result is determined from the first retrieval result and the second retrieval result.