Abstract: Described herein are systems, methods, and instrumentalities associated with landmark detection. The detection may be accomplished by determining a graph representation of a plurality of hypothetical landmarks detected in one or more medical images. The graph representation may include nodes that represent the hypothetical landmarks and edges that represent the relationships between paired hypothetical landmarks. The graph representation may be processed using a graph neural network such a message passing graph neural network, by which the landmark detection problem may be converted and solved as a graph node labeling problem.

Abstract: This application discloses a communication method and a communications apparatus. The method includes: receiving, by a terminal device, reference signal configuration information sent by a network device, where the reference signal configuration information includes at least one of the following information: a beam sweeping type and a reference signal beam indication; receiving, by the terminal device, a reference signal and data that are sent by the network device; and determining, by the terminal device based on the reference signal configuration information, whether data can be mapped onto another resource element that is located on a same orthogonal frequency division multiplexing (OFDM) symbol as the reference signal. The corresponding apparatus is further disclosed. According to technical solutions provided in this application, reliable data reception can be implemented.

Abstract: A guidance control device is configured to control a projection device for projecting images. The guidance control device is configured to generate a subject-identifying image to be projected on a first projection range relating to a first position at which a subject person conducts a predetermined operation and to generate a guide image guiding the predetermined operation for the subject person to be projected on a second projection range relating to a second position viewable by the subject person. Accordingly, the guidance control device is able to control the projection device including the subject-identifying image and the guide image.

Abstract: Example communication methods and apparatus are described. One example communication method includes generating a synchronization signal block (SSB) by an access network device, where the SSB includes a first physical broadcast channel (PBCH) and a second PBCH. The first PBCH carries time-frequency resource position information of first synchronization information, the second PBCH carries time-frequency resource position information of second synchronization information, and a time-frequency resource position of the first synchronization information is different from a time-frequency resource position of the second synchronization information. The access network device broadcasts the SSB. In embodiments of this application, the SSB may carry PBCHs corresponding to terminal devices of different bandwidth types and/or synchronization information of the terminal devices of different bandwidth types to satisfy processing requirements of the terminal devices of a plurality of bandwidth types.

Abstract: An electronic device includes a network monitor configured to acquire network environment information related to a radio frequency (RF) transmission signal; a transceiver configured to generate an envelope signal of the RF transmission signal; a transmission (Tx) module including a power amplifier for receiving the RF transmission signal from the transceiver and amplifying the RF transmission signal; and an envelope tracking (ET) modulator configured to receive the envelope signal from the transceiver and to provide a bias of a power amplifier to correspond to the envelope signal, wherein the ET modulator determines a magnitude of the bias of the power amplifier based on the network environment information acquired by the network monitor.

Abstract: The present disclosure provides systems and methods for processing a 3D image. A method may include for an orientation of the 3D image, obtaining, from the 3D image, a plurality of slices of the orientation; processing the plurality of slices of the orientation by applying a conversion model; and determining a converted 3D image of the orientation based on the plurality of processed slices of the orientation.

Abstract: A method is provided of training a generative adversarial network for performing semantic segmentation of images. The generative adversarial network includes a generator neural network and a discriminator neural network. The method includes providing an image as input to the generator neural network, receiving a predicted segmentation map for the image from the generator neural network, providing i) the image, ii) the predicted segmentation map, and iii) ground-truth label data corresponding to the image, as distinct training inputs to the discriminator neural network, determining a set of one or more outputs from the discriminator neural network in response to said training inputs, and training the generator neural network using a loss function that is a function of said set of outputs from the discriminator neural network.

Abstract: An intraoral scanning system comprises an intraoral scanner and a computing device. The computing device receives intraoral scans of a dental site from the intraoral scanner, the intraoral scans comprising at least a first intraoral scan and a second intraoral scan. The computing device determines, for each point in the second intraoral scan based on a first comparison of the first intraoral scan to the second intraoral scan, whether the point is a back-face point or an uncovered point. The computing device determines a region comprising a plurality of points in the second intraoral scan that are uncovered points, determines whether the region satisfies one or more removal criteria, and removes the region from the second intraoral scan responsive to determining that the region satisfies the one or more removal criteria.

Abstract: Various embodiments of a fall-detection system for detecting personal fall while preserving the privacy of a detected person are disclosed. This disclosed fall-detection system can begin by receiving a sequence of video images comprising a person being monitored. The disclosed fall-detection system then processes each video image in the sequence of video images by: detecting the person in the video image; and extracting a skeletal figure of the detected person by identifying a set of human keypoints from the detected person. Next, the disclosed fall-detection system processes the sequence of skeletal figures corresponding to the sequence of video images by labeling each skeletal figure in the sequence of skeletal figures with an action among a set of predetermined actions. The disclosed fall-detection system subsequently generates a fall/non-fall decision for the detected person based on the set of action labels corresponding to the sequence of video images.

Abstract: The present invention provides a object detection device for vehicle capable of performing distance measurement more reliably than in the past and having high compatibility for distance measurement. The present invention provides an object detection device 100 for vehicle including a first detection unit 10, a second detection unit 20, an error quantity calculation unit 30, a correction quantity calculation unit 40, and a distance correction unit 50. The error quantity calculation unit 30 compares the distances D1 and D1? of the same object detected by the first detection unit 10 and the second detection unit 20 to calculate the error quantity ?D. The correction quantity calculation unit 40 calculates the correction quantity CAt based on the error quantity ?D. The distance correction unit 50 corrects the distances D1? and D2? of the object detected by the second detection unit 20 based on the correction quantity CAt.

Abstract: A method for designing a dental component, specifically a restoration, a bite guard or an impression tray, comprising the steps of: measuring a dental situation; generating a 3D model of the dental situation; applying an artificial neural network for machine learning (convolutional neural network; CNN) to the 3D model of the dental situation and/or an initial 3D model of the dental component, and automatically generating a fully designed 3D model of the dental component, specifically the restoration, the bite guard or the impression tray.

Abstract: A computing device is provided, including a logic subsystem with one or more processors, and memory storing instructions executable by the logic subsystem. These instructions are executed to obtain one or more source images, segment the one or more source images to generate a plurality of segments, determine a priority order for the plurality of segments, and transmit the plurality of segments to a remote computing device in the priority order. The plurality of segments are spatial components generated by spatial decomposition of the one or more source images and/or frequency components that are generated by frequency decomposition of the one or more source images. A remote computing device may receive these components in priority order, and perform certain algorithms on individual components without waiting for the entire image to upload.

Abstract: An image processing device according to one embodiment estimates optical flow information, pixel by pixel, on the basis of a reference image and input images of consecutive frames, and estimates a term corresponding to temporal consistency between the frames of the input images. The image processing device determines a mesh on the basis of the term corresponding to temporal consistency and the optical flow information, and transforms the reference image on the basis of the mesh. The image processing device preforms image blending on the basis of the input image, the transformed reference image, and mask data.

Abstract: A decoding device, method, and computer readable medium for partial access metadata for video-based point cloud compression data are disclosed. The decoding device for point cloud decoding includes a communication interface and a processor operably coupled to the communication interface. The communication interface receives a compressed bitstream and metadata related to the compressed bitstream. The processor identifies, in the metadata, region information that indicates a region of a point cloud based on a user perspective; locates a track of the compressed bitstream that corresponds to the region information; identifies, in the track, objects corresponding to the region of the point cloud; and decodes for display of an object, included in the objects identified in the track, that is visible from the user perspective.

Abstract: This disclosure relates method and system for systematic disposal and dynamic procurement of recyclable waste resin. Typical systems for disposal and procurement of recyclable resin pose challenges such as lack of segregation and collection of different types of plastics, data on recycling centers in proximity, uncertainty over resin identification codes, and so on. The disclosed system provides a framework having multiple smart collection units that may be communicatively coupled with a server. The smart bins are capable of collecting plastic items and taking multiple images thereof. The system determines information on type, weight, and location of the plastic resin in the plastic item by means of a trained CNN model, and stores in a repository. The system further includes a route optimization model that is enables selection of a set of collection units for procurement of the resin (of a specific type and a specific quantity).

Abstract: Technologies for context-based management of wearable computing devices include a mobile computing device and a wearable computing device. The wearable computing device generates sensor data indicative of a location context of the wearable computing device and transmits the sensor data to the mobile computing device. The mobile computing device generates local sensor data indicative of a location context of the wearable computing device and fuses the local sensor data with the sensor data received from the wearable computing device. The mobile computing device determines a context of the wearable computing device based on the fused sensor data. The mobile computing device determines whether an adjustment to the functionality of the wearable computing device is required based on the determined context. The mobile computing device manages the functionality of the wearable computing device in response to determining that an adjustment to the functionality is required.

Abstract: An image is processed to determine semantic data about what is depicted in that image. The semantic data provides information about location in the image and semantic category of an object depicted within the image. The semantic categories may include “background”, “face”, “text”, “clothing”, “exercise”, and so forth. The semantic data may then be used to facilitate distribution of image data while limiting the information shared. For example, the semantic data may indicate the placement and position of a user exercising in first video data. Using the semantic data, second video data may be produced that obscures or replaces parts of the image that do not depict the user exercising. The second video data may then be used for providing feedback to the user, while preventing disclosure of other people, background, and so forth that are depicted in the first video data.

Abstract: Computer-implemented methods, program products, and systems for provenance-based defense against poison attacks are disclosed. In one approach, a method includes: receiving observations and corresponding provenance data from data sources; determining whether the observations are poisoned based on the corresponding provenance data; and removing the poisoned observation(s) from a final training dataset used to train a final prediction model. Another implementation involves provenance-based defense against poison attacks in a fully untrusted data environment. Untrusted data points are grouped according to provenance signature, and the groups are used to train learning algorithms and generate complete and filtered prediction models. The results of applying the prediction models to an evaluation dataset are compared, and poisoned data points identified where the performance of the filtered prediction model exceeds the performance of the complete prediction model.

Abstract: A method for detecting a fault in a solar panel in a solar farm is disclosed. In some examples, the method includes obtaining, by a computing system (106), a first image of the solar panel captured by an unmanned aerial vehicle (UAV) (102) during a first flight and a second image of the solar panel captured by the UAV (102) during a second flight. In some examples, the method also includes determining, by the computing system (106), a first score for the solar panel based on the first image and determining a second score for the solar panel based on the second image. In some examples, the method further includes determining, by the computing system (106), whether the solar panel has the fault based on the first score and the second score and outputting an indication that the solar panel has the fault in response to determining that the solar panel has the fault.

Abstract: A Kinect-based auxiliary training system for basic badminton movements, includes a data collection module, a movement feature extraction and recognition module, and a movement standard degree analysis and guidance module. The data collection module is provided with a Kinect v2 somatosensory device for monitoring athletes in real time, and collecting 3D coordinate data of 25 joint points of athletes' whole body. The movement feature extraction and recognition module is provided for establishing a standard template, and obtaining a similarity between the movement data and the standard template. The movement standard degree analysis and guidance module is provided for determining a category of the current movement of the user to be tested according to the similarity, and further analyzing whether the current movement of the user to be tested meets a standard according to a threshold range of the bone included angle set by a technology evaluation rule.