Abstract: A method, apparatus and computer program product provide an automated neural network training mechanism. The method, apparatus and computer program product receive a decoded noisy image and a set of input parameters for a neural network configured to optimize the decoded noisy image. A denoised image is generated based on the decoded noisy image and the set of input parameters. A denoised noisy error is computed representing an error between the denoised image and the decoded noisy image. The neural network is trained using the denoised noisy error and the set of input parameters and a ground truth noisy error value is received representing an error between the original image and the encoded image. The ground truth noisy error value is compared with the denoised noisy error to determine whether a difference between the ground truth noisy error value and the denoised noisy error is within a pre-determined threshold.

Abstract: Disclosed herein is a method for transmitting a wake-up signal in a wireless communication system supporting a wake-up radio (WUR) operation. The method performed by a network entity comprises transmitting a first wake-up signal for waking up a specific terminal to a specific access point (AP); transmitting a second wake-up signal to all APs within the coverage of the network entity if a response for the first wake-up signal is not received before a first timer expires; and transmitting, to a base station, a response message including control information indicating whether the specific terminal exists.

Abstract: The present disclosure provides systems and methods for virtual facial makeup simulation through virtual makeup removal and virtual makeup add-ons, virtual end effects and simulated textures. In one aspect, the present disclosure provides a method for virtually removing facial makeup, the method comprising providing a facial image of a user with makeups being applied thereto, locating facial landmarks from the facial image of the user in one or more regions, decomposing some regions into first channels which are fed to histogram matching to obtain a first image without makeup in that region and transferring other regions into color channels which are fed into histogram matching under different lighting conditions to obtain a second image without makeup in that region, and combining the images to form a resultant image with makeups removed in the facial regions.

Abstract: Interacting between users of mobile devices is provided by initiating a desired interaction at a first user device using a gesture sensed by the first user device, determining a gesture area, submitting a desired interaction request for interaction with associated users within the gesture area, and sending a notification to user devices indicating the desired interaction of the first user with one or more users of a local subset, wherein the notification identifies the first user with an identifier independent of identifiers used by the first user outside of interactions with a remote server.

Abstract: An apparatus for providing wireless communication between an RF communication device and remotely located RF base station devices or RF repeater devices to establish a wireless connection with a wireless carrier authorized to communicate with one or more wireless user devices (UEs) in communication with the RF communication device. The RF communication devices may employ separate wireless communication channels to communicate with one or more remote RF base station devices, remote RF repeater devices, remote network management applications, and local UEs. An RF communication device may employ a map to select a currently available RF base station device or an RF repeater device to establish a wireless connection between an authorized wireless carrier and the one or more local UEs in communication with the RF communication device.

Abstract: Iterative prediction systems and methods for the task of action detection process an inputted sequence of video frames to generate an output of both action tubes and respective action labels, wherein the action tubes comprise a sequence of bounding boxes on each video frame. An iterative predictor processes large offsets between the bounding boxes and the ground-truth.

Abstract: Systems and methods for detecting and tracking objects are provided. In one example, a computer-implemented method includes receiving sensor data from one or more sensors. The method includes inputting the sensor data to one or more machine-learned models including one or more first neural networks configured to detect one or more objects based at least in part on the sensor data and one or more second neural networks configured to track the one or more objects over a sequence of sensor data. The method includes generating, as an output of the one or more first neural networks, a 3D bounding box and detection score for a plurality of object detections. The method includes generating, as an output of the one or more second neural networks, a matching score associated with pairs of object detections. The method includes determining a trajectory for each object detection.

Abstract: Systems and methods to track the respective locations of subjects over time. The system identifies subjects who, over time, were co-located with one another suggesting they are associated with one another, and the pairs are analyzed. For each of the subjects, the system produces a vector that quantifies the subject's location history by including a respective weight for each combination of a time interval with a geographical area. The vectors are compared using a distance metric, and any pair of subjects whose vectors are sufficiently close are flagged as being an associated pair. The respective vector belonging to each subject is normalized to account for the total number of other subjects who were co-located with the subject. For each interval-area pair, the system may compute the frequency of the interval-area pair, and then divide each weight that corresponds to the interval-area pair by the frequency of the interval-area pair.

Abstract: A wireless telecommunication system comprises a base station and a terminal device (UE). The UE is configured to determine its mobility status and an associated reliability, e.g. an indication of the extent to which the UE is moving and the confidence in this determination. In some cases the reliability may be based on how the UE determined its mobility status. For example, if the UE determined itself to be of relatively low mobility because it has an internal configuration setting indicating it is deployed in a fixed position, the reliability of this mobility status may be considered to be greater than if the UE determined itself to be of relatively low mobility because it has not recently undergone many base station handovers. The UE is configured to convey an indication of the mobility status and the reliability to the base station.

Abstract: The present disclosure provides systems and methods for virtual facial makeup simulation through virtual makeup removal and virtual makeup add-ons, virtual end effects and simulated textures. In one aspect, the present disclosure provides a method for virtually removing facial makeup, the method comprising providing a facial image of a user with makeups being applied thereto, locating facial landmarks from the facial image of the user in one or more regions, decomposing some regions into first channels which are fed to histogram matching to obtain a first image without makeup in that region and transferring other regions into color channels which are fed into histogram matching under different lighting conditions to obtain a second image without makeup in that region, and combining the images to form a resultant image with makeups removed in the facial regions.

Abstract: A target object selection method, a server, and a computer storage medium are disclosed. The method includes selecting a first one candidate user meeting a first condition in a target area, the first condition characterizing a user who has appeared in the target area; extracting a first social relationship chain of the at least one candidate user in the target area, the social relationship chain comprising a first related user who has a social relationship with the candidate user; and selecting a first one target user in the target area based on the first social relationship chain and sending information to the first target user.

Abstract: The present disclosure provides systems and methods for virtual facial makeup simulation through virtual makeup removal and virtual makeup add-ons, virtual end effects and simulated textures. In one aspect, the present disclosure provides a method for virtually removing facial makeup, the method comprising providing a facial image of a user with makeups being applied thereto, locating facial landmarks from the facial image of the user in one or more regions, decomposing some regions into first channels which are fed to histogram matching to obtain a first image without makeup in that region and transferring other regions into color channels which are fed into histogram matching under different lighting conditions to obtain a second image without makeup in that region, and combining the images to form a resultant image with makeups removed in the facial regions.

Abstract: A mechanism for controlling secondary-coverage scanning for configuring dual-connectivity. A computing system establishes for a primary access node a scan list of secondary cells based on the primary access node having an X2 interface for each secondary cell and having received a threshold extent of B1 measurement reporting indicating threshold strong coverage of the secondary cell for configuration of dual-connectivity. Further, the computing system establishes for each secondary cell of the scan list a respective B1 measurement threshold based on evaluation of performance history of the secondary cell such as based on extent of secondary-access-node-addition failures involving the secondary cell and/or based on communication-quality metrics such as error rate or retransmission rate. The computing system then configures the primary access node to provide a B1 measurement object conveying the scan list with measurement thresholds.

Abstract: A method and apparatus are provided for using a neural network to estimate scatter in X-ray projection images and then correct for the X-ray scatter. For example, the neural network is a three-dimensional convolutional neural network 3D-CNN to which are applied projection images, at a given view, for respective energy bins and/or material components. The projection images can be obtained by material decomposing spectral projection data, or by segmenting a reconstructed CT image into material-component images, which are then forward projected to generate energy-resolved material-component projections. The result generated by the 3D-CNN is an estimated scatter flux. To train the 3D-CNN, the target scatter flux in the training data can be simulated using a radiative transfer equation method.

Abstract: A method for the recognition of a moving pedestrian by means of a radar sensor system includes the steps of transmitting a primary radar signal into an observation space and of receiving a secondary radar signal reflected by the moving pedestrian. The secondary radar signal is processed, wherein the processing of the secondary radar signal includes the steps of generating a Micro-Doppler spectro-gram of the secondary radar signal, determining, based on the Micro-Doppler spectrogram, an observed bulk speed of the moving pedestrian, determining, based on the Micro-Doppler spectrogram, at least one gait cycle parameter of the moving pedestrian and determining, based on the determined observed bulk speed and the determined gait cycle parameter, an illumination angle between a moving direction of the moving pedestrian and the line of sight.

Abstract: A method for determining volume of an object includes: obtaining a target depth image containing a target object which is captured by a depth image capturing device (S101); performing segmentation based on depth data in the target depth image to obtain a target image region corresponding to the target object (S102); determining a target circumscribed rectangle that corresponds to the target image region and meets a predetermined condition (S103); determining volume of the target object based on the target circumscribed rectangle and the depth data of the target depth image (S104). Compared to the laser-based determination method in the prior art, this method employs a depth image capturing device without a laser measurement device, and thus the economic cost is lower. Further, compared to the manual scale-based determination method in the prior art, the method employs a software program to automatically determine volume without manual cooperation, and has a higher accuracy and efficiency.

Abstract: A perception device, including at least one image sensor configured to detect a plurality of images; an information estimator configured to estimate from each image of the plurality of images a depth estimate, a velocity estimate, an object classification estimate and an odometry estimate; a particle generator configured to generate a plurality of particles, wherein each particle of the plurality of particles comprises a position value determined from the depth estimate, a velocity value determined from the velocity estimate and a classification value determined from the classification estimate; an occupancy hypothesis determiner configured to determine an occupancy hypothesis of a predetermined region, wherein each particle of the plurality of particles contributes to the determination of the occupancy hypothesis.

Abstract: The present disclosure provides a method and apparatus for evaluating image definition, a computer device and a storage medium, wherein the method comprises: obtaining an image to be processed; inputting the image to be processed to a pre-trained evaluation model; obtaining an comprehensive image definition score outputted by the evaluation model, the comprehensive image definition score being obtained by the evaluation model by obtaining N image definition scores based on N different scales respectively, and then integrating the N image definition scores, N being a positive integer greater than one. The solution of the present invention can be applied to improve the accuracy of the evaluation result.

Abstract: An electronic device includes: a display having a notch portion formed on at least one edge thereof, a processor; and a memory configured to store at least one of a position, a size, or a color of a region corresponding to the notch portion, wherein the memory is configured to store instructions that cause the processor to identify an image characteristic of a non-display region of an image that is not displayed by the region corresponding to the notch portion when receiving a user command requesting displaying of the image, and to control the display according to the image characteristic to adjust a region displaying the image and to display the image. Other embodiments are possible.

Abstract: A system for real-time depth sensing includes a structured-light stereo device, a time of flight device, and a computing device. The computing device executes instructions that cause the computing device to determine depth measurements of a scene using information received by the structured-light stereo device and to determine time of flight measurements of the scene using information received by the time of flight device. The computing device executes further instructions that cause the computing device to generate a depth map using the depth measurements, to generate calibration points using the time of flight measurements, and to update the depth map using the calibration points.