Abstract: A face attribute recognition method, electronic device, and storage medium. The method may include obtaining a face image, inputting the face image into an attribute recognition model, performing a forward calculation on the face image using the attribute recognition model to obtain a plurality of attribute values according to different types of attributes, and outputting the plurality of attribute values, the plurality of attribute values indicating recognition results of a plurality of attributes of the face image. The attribute recognition model may be obtained through training based on a plurality of sample face images, a plurality of sample attribute recognition results of the plurality of sample face images, and the different types of attributes.

Abstract: A method and system may be used to determine whether a suspect item is counterfeit. In the method and system, images may be analyzed in order to identify regions of interest of the item. The method and system may compare characteristics of different regions of interests against threshold characteristics derived from an item model. A determination that the suspect item is counterfeit is based on a result of the comparison.

Abstract: An anomaly detection apparatus 100 includes an image transformation unit 103 that calculates an image transformation parameter, based on an inspection image in which an inspection object appears, a reference image indicating a normal state of the inspection object and a parameter for image transformation parameter calculation, and performs image transformation on the inspection image using the image transformation parameter, an image change detection unit 104 that collates the reference image and the image-transformed inspection image using a change detection parameter, and calculates an anomaly certainty factor indicating whether there is a change in a specific region of the inspection image, a change detection parameter learning unit 106 that learns the change detection parameter, based on a difference between a training image indicating a correct answer value of the change and the anomaly certainty factor, and an image transformation parameter learning unit 108 that learns the parameter for image transform

Abstract: A method includes receiving sensor data indicative of a paved surface, and identifying a defect associated with the paved surface based at least in part on the sensor data. The method also includes determining that the defect is of a defect type based on determining that a value associated with the defect is within a value range associated with the defect type. The method further includes generating a command associated with the defect that, when executed by a machine, at least partially remedies the defect. The method also includes providing the command to an electronic device via a network.

Abstract: A machine learning system extracts features from images. A system and method predicts appearance type, attractiveness, gender, or ethnicity from a facial image. In a machine learning approach, predictor training is performed by using known data sets containing the classification information for a given facial image into the desired categories. When a new image is presented, the system applies this info to the new image. The training can be performed continuously while the system is being presented with new images and corrective results in a feedback loop. In another machine learning approach, a generative adversarial network is used to predict the desired categories based on the given facial image.

Abstract: The present disclosure provides an image processing method applied to a vehicle head-up display device, including: determining position information of a target object within a range of activity; acquiring a distortion parameter corresponding to the position information by looking up a table; and performing distortion correction processing on an image to be displayed according to the distortion parameter, so as to obtain a distortion corrected image.

Abstract: The present disclosure provides an image segmentation apparatus, method and relevant computing device. The image segmentation apparatus comprises: a feature extractor configured to extract N image semantic features having different scales from an input image, where N is an integer not less than 3; and a feature processor comprising cascaded dense-refine networks and being configured to perform feature processing on the N image semantic features to obtain a binarized mask image for the input image. A dense-refine network is configured to generate a low-frequency semantic feature from semantic features input thereto by performing densely-connected convolution processing on the semantic features respectively to obtain respective image global features, performing feature fusion on the image global features to obtain a fused image global feature, and performing pooling processing on the fused image global feature to generate and output the low-frequency semantic feature.

Abstract: Implementations are described herein for automatically generating synthetic training images that are usable as training data for training machine learning models to detect, segment, and/or classify various types of plants in digital images. In various implementations, a digital image may be obtained that captures an area. The digital image may depict the area under a lighting condition that existed in the area when a camera captured the digital image. Based at least in part on an agricultural history of the area, a plurality of three-dimensional synthetic plants may be generated. The synthetic training image may then be generated to depict the plurality of three-dimensional synthetic plants in the area. In some implementations, the generating may include graphically incorporating the plurality of three-dimensional synthetic plants with the digital image based on the lighting condition.

Abstract: An apparatus and method for training a classification model and an apparatus for classifying with a classification model are disclosed. The apparatus for training a classification model comprises: a local area obtainment unit to, obtain predetermined local area of each sample image; a feature extraction unit to, with respect to each sample image, set corresponding numbers of feature extraction layers for the global area and each predetermined local area, to extract a global feature of the global area and a local feature of each predetermined local area, wherein the global area and the predetermined local areas share at least one feature extraction layer in which the global feature and each local feature are combined; and a loss determination unit to calculate a loss function of the sample image based on combined features of each sample image, and to train the classification model based on the loss function.

Abstract: Embodiments of the present disclosure provide a scheduling method and a base station. In a cell handover process, a target base station serving a target cell receives a random access request from a terminal, sends a random access response to the terminal according to the random access request, and proactively sends an uplink grant to the terminal after sending the random access response and prior to receiving a handover complete command from the terminal. By proactively sending the uplink grant, the target base station can perform uplink scheduling for UE before cell handover is completed, so as to reduce an uplink transmission latency caused by the cell handover.

Abstract: Obstacle determination systems and programs detect back-and-forth steering that is behavior of a vehicle for avoiding an obstacle. The systems and programs determine that the obstacle is present between a point where the back-and-forth steering has started and a point where the back-and-forth steering has ended when a steering start road and a steering end road are identical roads. The steering start road is a road where the vehicle is present at a start of the back-and-forth steering. The steering end road is a road where the vehicle is present at an end of the back-and-forth steering.

Abstract: Described herein is a method for image processing in which a captured image is processed to provide a background image which is updated in accordance with relative movement between foreground and background in successive captured images.

Abstract: A system for detecting flying animals, the system comprising: a detection module configured to detect flying animals; a local processing module associated with the detection module and configured to remove background noise from data gathered by the detection module; an analysis module configured to receive data from the local processing module and process said data to determine presence of flying animals and classify such flying animals by species; and a deterrent configured to repel one or more species of flying animal based on the classification of detected flying animals.

Abstract: A method, apparatus, and computer program product are provided for determining a location of on-street parking. The method includes receiving a plurality of individual probe points along a section of a roadway. Each individual probe point is associated with a location where each probe point was captured by one of one or more vehicles on the section of the roadway. The method also includes selecting one or more of the plurality of individual probe points based on a speed of the one or more vehicles associated with each probe point at a time the probe point was captured. The method further includes comparing the location of the selected probe points with a known roadway representation. Based on the comparison, the method also includes determining one or more groupings of a plurality of potential parking spaces along the section of the roadway. A corresponding apparatus and computer program product are provided.

Abstract: A method and an apparatus for handling in-device coexistence interference in a user equipment are provided. The method includes receiving a list of non-serving Long Term Evolution (LTE) frequencies from a base station, configuring the non-serving LTE frequencies so as to perform a measurement on the configured non-serving LTE frequencies, and detecting a likelihood of in-device coexistence interference between at least one of the non-serving LTE frequencies and an Industrial, Scientific and Medical (ISM) frequency when an ISM activity is ongoing on the ISM frequency.

Abstract: In accordance with some embodiments, an apparatus for privacy protection includes a housing arranged to hold a personal communication device. The apparatus further includes at least one off-device resource supported by the housing that provides an alternate to a corresponding on-device resource integrated within the personal communication device. The apparatus also includes a local communication device coupled to the at least one off-device resource and supported by the housing, where the local communication device includes a personal communication device interface modem operable to provide a communication channel between the at least one off-device resource and the personal communication device.

Abstract: Systems and methods for object recognition with limited input are disclosed herein. An example method includes updating a neural network trained to perform object recognition on a first rendition of an object, so that the neural network performs object recognition on a second rendition of the object, using a limited set of input images. The method includes receiving a limited set of model images of the second rendition of the object, accessing a corresponding image mapping, and generating a large number of training images from the limited set, where image mappings include geometric, illumination, and/or obscuration transformations. The neural network is then trained, from this initial small set, to classify the second rendition of the object.

Abstract: Systems, methods, and computer readable mediums for configuring a radio system to provide network service access based on a location of the radio system. The radio system can identify the location and send a location signal indicating its location over a first network. The radio system can receive, over the first network, first configuration data for communicating over a second network based on its location. The radio system can configure itself to communicate over the second network using the first configuration data and subsequently provide network service access over the second network according to the first configuration data.

Abstract: An artificial intelligence (AI) system which utilizes a machine learning algorithm such as deep learning, a method therefor and an application therefor are provided. The method includes controlling an electronic device provided in a vehicle of the disclosure includes obtaining region information of a region located on a periphery of a path of the vehicle, obtaining an image by photographing a road on which the vehicle travels, through a camera, obtaining road information regarding the road on which the vehicle travels, by analyzing the obtained image, obtaining information regarding a virtual danger object which is likely to appear on the road on which the vehicle travels, based on the road information and the region information, and providing the information regarding the virtual danger object.

Abstract: An SMS-based VoWIFI configuration method comprises: when an SMS of a VoWIFI type is received, extracting a configuration identifier carried by the SMS, and determining the type of the configuration identifier; and when the configuration identifier is an enable identifier, extracting EPDG information carried by the SMS, and registering an IMS service according to the EPDG information, so as to enable a VoWIFI function.