Devices, methods, and computer-readable media are disclosed, describing an adaptive, subject-aware approach for image bracket selection and fusion, e.g., to generate high quality images in a wide variety of capturing conditions, including low light conditions. An incoming image stream may be obtained from an image capture device, comprising images captured using differing default exposure values, e.g., according to a predetermined pattern. When a capture request is received, it may be detected whether one or more human or animal subjects are present in the incoming image stream. If a subject is detected, an exposure time of one or more images selected from the incoming image stream may be reduced relative to its default exposure time. Prior to the fusion operation, one of the selected images may be designated a reference image for the fusion operation based, at least in part, on a sharpness score and/or a blink score of the image.

A method and apparatus for generating a face-harmonized image are disclosed. The method of generating a face-harmonized image includes receiving an input image, extracting facial landmarks from a target image and the input image, generating a face-removed image of the target image based on a facial mask region, extracting a user face image from the input image, transforming the user face image to correspond to the facial mask region, generating a face-blended image by blending the transformed user face image with the target image, extracting a feature map of the face-blended image, generating a combined feature map based on the feature map of the face-blended image and a feature map of the target image, generating a face harmonization result image based on the combined feature map, and providing the generated face harmonization result image.

The present invention provides a robust and effective solution to an entity or an organization by enabling them to implement a system for swapping one or more faces without any explicit training on the one or more faces. The proposed method can be further implemented in real time.

A system and a method of performing face recognition may include: receiving a first facial image, depicting a first face, and a second facial image depicting a second face; applying an ML model on the first image, to produce a first representation vector, and applying the ML model on the second image to produce a second representation vector; comparing the first representation vector and the second representation vector; and associating the first face with the second face based on the comparison, where the ML model is trained to produce the representation vectors from the facial images, based on regions in the facial images that correspond to distinctiveness scores that are beneath a distinctiveness threshold.

A gate apparatus includes an acquisition unit, a matching request unit, and a control unit. The acquisition unit acquires biological information about an examination target user. The matching request unit requests a server apparatus that stores biological information about users and MRZ (Machine Readable Zone) information written in machine readable zones in passports issued to the users in association with each other to perform matching on the biological information about the examination target user. The control unit controls a gate so that the examination target user can pass through the gate if information is successfully read out from an IC (Integrated Circuit) chip in a passport issued to the examination target user by using MRZ information determined by the matching.

Disclosed are a face image and iris image acquisition method and device, a computer-readable readable storage medium and an apparatus. The method includes rotating the first tripod head to force the face lens and the iris lens to be in acquisition positions; capturing a first face image and a first iris image simultaneously by the face lens and the iris lens; and locating the iris in the first iris image, and if no iris is located, determining whether a condition of light-avoiding rotation is satisfied, and if the condition is satisfied, rotating the second tripod head to adjust an angle or a position of the supplementary light source to enable a light spot region to avoid an iris region.

Embodiments of the present disclosure provide a quantitative method and system for attention based on a line-of-sight estimation neural network, which improves the stability and training efficiency of the line-of-sight estimation neural network. A few-sample learning method is applied to training of the line-of-sight estimation neural network, which improves generalization performance of the line-of-sight estimation neural network. A nonlinear division method for small intervals of angles of the line of sight is provided, which reduces an estimation error of the line-of-sight estimation neural network. Eye opening and closing detection is added to avoid the line-of-sight estimation error caused by an eye closing state. A method for solving a landing point of the line of sight is provided, which has high environmental adaptability and can be quickly used in actual deployment.

Disclosed is an electronic device. The electronic device comprises: a microphone comprising circuitry; a speaker comprising circuitry; and a processor electrically connected to the microphone and speaker, wherein the processor, when a first user's voice is input through the microphone, identifies a user who uttered the first user's voice and provides a first response sound, which is obtained by inputting the first user's voice to an artificial intelligence model learned through an artificial intelligence algorithm, through the speaker, and when a second user's voice is input through the microphone, identifies a user who uttered the second user's voice, and if the user who uttered the first user's voice is the same as the user who uttered the second user's voice, provides a second response sound, which is obtained by inputting the second user's voice and utterance history information to the artificial intelligence model, through the speaker. In particular, at least some of the methods of providing a response sound to a user's voice may use an artificial intelligence model learned in accordance with at least one of a machine learning, neural network, or deep learning algorithm.

A backlight face recognition method, a terminal device using the same, and a computer readable storage medium are provided. The method includes: performing a face detection on each original face image in an original face image sample set to obtain a face frame corresponding to the original face image; capturing the corresponding original face images from the original face image sample set, and obtaining a new face image containing background pixels corresponding to the captured original face images from the original face image sample set; preprocessing all the obtained new face images to obtain a backlight sample set and a normal lighting sample set; and training a convolutional neural network using the backlight sample set and the normal lighting sample set until the convolutional neural network reaches a preset stopping condition. The trained convolutional neural network will improve the accuracy of face recognition in complex background and strong light.

Response methods to human abnormal behaviors for a mobility aid robot having a user-facing camera are disclosed. The mobility aid robot responds to human abnormal behaviors by detecting a face of a human during the robot aiding the human to move through the camera, comparing an initial size of the face and an immediate size of the face in response to the face of the human having detected during the robot aiding the human to move, determining the human as in abnormal behavior(s) in response to the immediate size of the face being smaller than the initial size of the face, and performing response(s) corresponding to the abnormal behavior(s) in response to the human being in the abnormal behavior(s), where the response(s) include slowing down the robot.