Abstract: Methods and systems are provided for generating mattes for input images. A neural network system is trained to generate a matte for an input image utilizing contextual information within the image. Patches from the image and a corresponding trimap are extracted, and alpha values for each individual image patch are predicted based on correlations of features in different regions within the image patch. Predicting alpha values for an image patch may also be based on contextual information from other patches extracted from the same image. This contextual information may be determined by determining correlations between features in the query patch and context patches. The predicted alpha values for an image patch form a matte patch, and all matte patches generated for the patches are stitched together to form an overall matte for the input image.

Abstract: The present application relates to an image processing method and system. The method includes: determining an enhanced image of a target object of an input image based on a segmentation algorithm, where the enhanced image of the target object comprises an image in which each pixel classified as the target object is displayed in an enhanced manner; and determining a positioning image of the target object by applying an integral image algorithm to the enhanced image of the target object.

Abstract: Some methods enable a computer to receive location information from a device and determine a coverage region for the device. The device's coverage region is the area within range of a wireless signal coming from the device. The computer can determine that part of the device's coverage region covers part of a specified region and instruct the device to provide wireless access to a network associated with the first region.

Abstract: The present invention provides a motion capture with a high accuracy which can replace an optical motion capture technology, without attaching optical markers and sensors to a subject. A subject with an articulated structure has a plurality of feature points in the body of the subject including a plurality of joints wherein a distance between adjacent feature points is obtained as a constant. A spatial distribution of a likelihood of a position of each feature point is obtained based on a single input image or a plurality of input images taken at the same time. One or a plurality of position candidates corresponding to each feature point are obtained based on the spatial distribution of the likelihood of the position of each feature point. Each join angle is obtained by performing an optimization calculation based on inverse kinematics using the candidates and the articulated structure.

Abstract: High dynamic range (HDR) support is provided for legacy application programs, such as games that are configured to display standard dynamic range (SDR) frames. HDR frames may be synthesized without modifying the legacy application program. The buffer creation process of the legacy application program is intercepted and modified before creation of the SDR format buffer so that the buffer is configured to use an HDR format. A location of an intermediate buffer storing HDR rendered data is determined by intercepting and analyzing graphics driver calls in a command stream produced by the legacy application program. The HDR rendered data is combined with user interface content extracted from the SDR frames. Additionally, any post processing effects used by the legacy application program to produce the SDR frames may be predicted and applied to the HDR rendered data to synthesize the HDR frames for display on a modern HDR display device.

Abstract: An integrated system for processing and using images acquired of subjects in a research or clinical environment is provided. The integrated system includes an image and data bank, a workflow management module, a cloud storage module and a pre-processing module. The pre-processing engine is positioned between the workflow management module and the cloud storage module and is configured to receive raw images, data associated with the raw images and the data associated with the imaged subjects from the workflow management module and process this data and provide the processed data to be pushed into the cloud storage module. The pre-processing engine also anonymizes the data to provide de-identified images and data to the cloud storage module and creates a key that relates the raw images, data associated with the raw images and the data associated with the imaged subjects to the de-identified, processed images and data. The key remains separate and un-connected from the de-identified, processed images and data.

Abstract: The present application provides a model training method and apparatus, an electronic device and a readable storage medium. The model training method of a pedestrian re-identification network, includes: determining a current weight matrix for multiple identification objects according to a current training result of a neural network model; calculating a difficulty vector corresponding to the multiple identification objects according to the current weight matrix; and, selecting next training samples of the multiple identification objects for a next training according to the difficulty vector.

Abstract: A method for generating result slice images with at least partially different slice thickness based on a tomosynthesis image data set of a breast includes generating average value slices and maximum value slices (MIP) based on the tomosynthesis image data set, frequency dividing the average value slices into low-pass filtered and high-pass filtered average value slices, high-pass filtering of maximum value slices to form high-pass filtered maximum value slices, mixing high-pass filtered maximum value slices and high-pass filtered average value slices to form mixed high-pass filtered maximum value slices, combining the low-pass filtered average value slices with the mixed high-pass filtered maximum value slices to form the result slice images, and applying a moving maximum value across a selected thickness of maximum value slices or across a selected thickness of mixed high-pass filtered maximum value slices.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately, efficiently, and flexibly generating harmonized digital images utilizing a self-supervised image harmonization neural network. In particular, the disclosed systems can implement, and learn parameters for, a self-supervised image harmonization neural network to extract content from one digital image (disentangled from its appearance) and appearance from another from another digital image (disentangled from its content). For example, the disclosed systems can utilize a dual data augmentation method to generate diverse triplets for parameter learning (including input digital images, reference digital images, and pseudo ground truth digital images), via cropping a digital image with perturbations using three-dimensional color lookup tables (“LUTs”).

Abstract: Measurement resources are often limited in 5G and 4G-LTE communications. As a result, SMTC and CSI-RS measurement windows will occasionally be assigned to the same resource elements of a signal frame. This results in a collision that must be resolved, since both measurements cannot be performed simultaneously. The collision can be resolved using restriction, where overlapping resource elements are assigned to either the SMTC or the CSI-RS measurements, or can be resolved using measurement sharing, where overlapping resource elements are divided among the two measurements according to a measurement sharing factor. Various factors, including priority or window periodicity may be taken into account by the measurement rules that dictate collision resolution.

Abstract: A convolutional neural network system includes a sensor and a controller, wherein the controller is configured to receive an image from the sensor, divide the image into patches, each patch of size p, extract, via a first convolutional layer, a feature map having a number of channels based on a feature detector of size p, wherein the feature detector has a stride equal to size p, refine the feature map by alternatingly applying depth-wise convolutional layers and point-wise convolutional layers to obtain a refined feature map, wherein the number of channels in the feature map and the size of the feature map remains constant throughout all operations in the refinement; and output the refined feature map.

Abstract: This application relates to a method and an apparatus for detecting facial key points, a computer device, and a storage medium including: acquiring a to-be-detected face image from a current frame; determining partial images in the to-be-detected face image, each partial image including one or more facial key points; determining, within each of the partial images, candidate points of the one or more facial key points corresponding to the partial image, respectively; and jointly constraining the candidate points in the partial images to determine a set of facial key points from the candidate points for the to-be-detected face image. For the partial images in the entire to-be-detected face image, the candidate points of the facial key points corresponding to the partial images are respectively determined. Therefore, a calculation amount may be reduced, and the efficiency of determining the candidate points of the facial key points is improved.

Abstract: The present disclosure discloses an Internet of Things (IoT) system for industrial data processing, a control method, and a storage medium, and provides a technical solution for a timely adjustment of production line parameters according to image information generated during a machine vision data collection on the production line. Through a difference of the image information corresponding to different process operations, a processing situation of different process operations may be obtained, so that more accurate adjustment of the production line parameter may be achieved without increasing the system complexity, thereby effectively reducing a development cost of the IoT, and increasing accuracy of the intelligent manufacturing control.

Abstract: According to one embodiment, a reading system includes an extractor, a determiner, and a reader. The extractor extracts a candidate image from an input image. The candidate image is a candidate of a portion of the input image in which a segment display is imaged. The determiner uses the candidate image and a mask to calculate a match ratio indicating a certainty of a segment display being included in the candidate image, and determines that the candidate image is an image of a segment display when the match ratio is not less than a threshold. The mask and the threshold are preset. The reader reads a numerical value displayed in a segment display from the candidate image determined to be an image of a segment display.

Abstract: Example aspects include a method, an apparatus and a computer-readable medium of enforcing a mask wearing rule, comprising monitoring video frames of one or more video feeds. The aspects further include detecting a person in at least one video frame of the video frames. The at least one video frame corresponding to a first area of one or more areas. Additionally, the aspects further include determining whether the person is in violation of a mask wearing rule. Additionally, the aspects further include generating an alert in response to determining that the person is in violation of the mask wearing rule.

Abstract: The present disclosure describes image analysis techniques that identify the source of a document. Once the source of the document is determined, the image analysis may locate one or more anchor fields in the document. The anchor fields may identify one or more additional fields that contain time-sensitive data and/or information. The image analysis performed herein may identify the time-sensitive data and/or information and process the data and/or information to schedule due dates and reminders.

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: The present disclosure provides a method and an apparatus for semantic segmentation of an image, capable of solving the problem in the related art associated with low speed and inefficiency in semantic segmentation of images. The method includes: receiving the image; performing semantic segmentation on the image to obtain an initial semantic segmentation result; and inputting image information containing the initial semantic segmentation result to a pre-trained convolutional neural network for semantic segmentation post-processing, so as to obtain a final semantic segmentation result. With the solutions of the present disclosure, the initial semantic segmentation result can be post-processed using the convolutional neural network, such that the speed and efficiency of the semantic segmentation of the image can be improved.

Abstract: The present disclosure generally relates to systems that include an artificial intelligence (AI) architecture for determining whether an image is manipulated. The architecture can include a constrained convolutional layer, separable convolutional layers, maximum-pooling layers, a global average-pooling layer, and a fully connected layer. In one specific example, the constrained convolutional layer can detect one or more image-manipulation fingerprints with respect to an image and can generate feature maps corresponding to the image. The global average-pooling layer can generate a vector of feature values by averaging the feature maps. The fully connected layer can then generate, based on the vector of feature values, an indication of whether the image was manipulated or not manipulated.

Abstract: The present disclosure describes methods and systems, including computer-implemented methods, computer program products, and computer systems, for monitoring the flow of drilling mud from a wellbore by image processing. One method includes: capturing, using a digital imaging device, images of drilling mud at a surface of the wellbore as the drilling mud flows through one or more image capture zones of a circulation system that circulates drilling mud through the wellbore and a wellbore drilling assembly; receiving, by one or more processors of a computer system operatively coupled to the digital imaging device, the images captured by the digital imaging device; and processing, by the one or more processors, the images captured by the digital imaging device to determine a rate of flow of the drilling mud through the one or more image capture zones.