Abstract: An edge computing system is deployed at a physical location and receives an input from one or more image/video sensing mechanisms. The edge computing system executes artificial intelligence image/video processing modules on the received image/video streams and generates metrics by performing spatial analysis on the images/video stream. The metrics are provided to a multi-tenant service computing system where additional artificial intelligence (AI) modules are executed on the metrics to execute perception analytics. Client applications can then be run on the output of the AI modules in the multi-tenant service computing system.

Abstract: A deposit detection device according to an embodiment includes an adhesion detection module, a moving adhesion detection module, and a determination module. The adhesion detection module detects a deposit region corresponding to a deposit adhering to an imaging device, based on brightness information of an image captured by the imaging device. The moving adhesion detection module detects the deposit region detected during moving of the vehicle as a moving deposit region, from among the deposit regions detected by the adhesion detection module. When the area of the moving deposit region detected by the moving adhesion detection module is equal to or larger than a first threshold value, the determination module determines that there is a deposit.

Abstract: An electronic device and a controlling method thereof are provided. A controlling method of an electronic device according to the disclosure includes: performing first learning for a neural network model for acquiring a video sequence including a talking head of a random user based on a plurality of learning video sequences including talking heads of a plurality of users, performing second learning for fine-tuning the neural network model based on at least one image including a talking head of a first user different from the plurality of users and first landmark information included in the at least one image, and acquiring a first video sequence including the talking head of the first user based on the at least one image and pre-stored second landmark information using the neural network model for which the first learning and the second learning were performed.

Abstract: A system for physical-virtual environment fusion includes a sensor with a computing system and a memory in communication with the computing system, the memory storing a plurality of endpoints. The computing system is configured to select a semantic identity of an object from among a group of objects indicated by a user pointer indicator by determining, based on inputs from the sensor, the approximate orientation of the user pointer indicator towards a first endpoint from among the plurality of endpoints, the first endpoint having the group of objects located within the first endpoint. It further renders, on a display surface, the group of objects, receives from the user an indication regarding the semantic identity of the object and presents, on the display surface, an indication of the selection of the object from among the group of objects based on the further indications from the user.

Abstract: A scene change method and system combining instance segmentation and cycle generative adversarial networks are provided. The method includes: processing a video of a target scene and then inputting the video into an instance segmentation network to obtain segmented scene components, that is, obtain mask cut images of the target scene; and processing targets in the mask cut images of the target scene by using cycle generative adversarial networks according to the requirements of temporal attributes to generate data in a style-migrated state, and generating style-migrated targets with unfixed spatial attributes into a style-migrated static scene according to a specific spatial trajectory to achieve a scene change effect.

Abstract: A video detection device includes a video acquisition interface, a video recognition interface, and a tilt angle estimation interface. The video acquisition interface acquires a video of the face of an imaging subject. The video recognition interface recognizes the left and right eyes from the video acquired by the video acquisition interface. The tilt angle estimation interface estimates a tilt angle with respect to the horizontal based on the position of the left and right eyes recognized by the video recognition interface.

Abstract: A system and method for operating a Vehicle to Infrastructure (V2I) system. The method includes receiving images from an infrared (IR) camera, determining whether a non-uniform noise exists within the received images, performing a calibration, upon determining that the non-uniform noise exists, performing a Non-Uniformity Correction (NUC) on the IR, upon determining that there is residual non-uniform noise, after performing the calibration, and determining that the IR camera has not detected a moving object that is approaching the IR camera, determining whether a Field of View (FOV) of the IR camera is occluded, after performing the NUC, and cleaning the IR camera, upon determining that the FOV of the infrared camera is occluded.

Abstract: This application discloses a communication method, a communications apparatus, a network device, and a terminal. Quasi co-location QCL beam information is determined. The QCL beam information includes beam information that has a QCL relationship with a reference signal antenna port. The QCL beam information is sent. A terminal receives the QCL beam information. The QCL beam information includes the beam information that has the QCL relationship with the reference signal antenna port. The terminal determines, based on the beam information that is included in the QCL beam information and that has the QCL relationship with the reference signal antenna port, a beam that has the QCL relationship with the reference signal antenna port.

Abstract: Methods and systems for vehicle passenger detection, can involve extracting a region of interest from one or more images of a vehicle captured by one or more cameras, image-processing of the region of interest and detecting faces in the region of interest with a pruned deep neural-network based object-detection module of a neural network comprising a pruned network, and utilizing the pruned network for inference to determine a number of passengers in the vehicle. The neural network can be pruned by identifying filter pairs in the neural network having a high correlation of weights to detect features have redundant features, and iteratively removing the filter pairs wherein the neural network is retrained after the iterative removal of the filter pairs.

Abstract: A system for classifying an object may include one or more processors, a sensor and a memory device. The memory device may include a data collection module, a starburst module, and an object classifying module. The modules have instructions that when executed by the one or more processors cause the one or more processors to obtain three dimensional point cloud data from the sensor, identify at least one cluster of points representing the object within the three dimensional point cloud data, identify a center point of the at least one cluster of points, project a plurality of rays from the center point to points of the at least one cluster of points to generate a shape, compare the shape to a plurality of candidate shapes, and classify the object when the shape matches at least one of the plurality of candidate shapes.

Abstract: Certain aspects of the present disclosure provide techniques for link configuration-specific modulation and coding scheme (MCS). A method that may be performed by a user equipment (UE) includes receiving control information indicating an MCS index corresponding to one or more MCS parameters to be used for a communication link, determining the one or more MCS parameters corresponding to the MCS index, wherein a mapping of the MCS index to the one or more MCS parameters is dependent on a configuration of the communication link, and communicating via the communication link in accordance with the one or more MCS parameters.

Abstract: The invention relates to a method and a device for determining at least one mechanical property of at least one object, wherein the at least one object is subjected to at least one interaction which influences a movement state of the at least one object, wherein the at least one interaction is selected such that the movement state effectuated by the interaction is a function of the at least one mechanical property, wherein the at least one object is observed using at least one camera, wherein the movement state is determined from images recorded by the at least one camera and wherein the at least one mechanical property of the at least one object is determined from the movement state thus determined.

Abstract: A method for processing a stream of data represented by a plurality of symbols associated with a constellation. The method includes applying a symbol constellation extension projection to at least one of the plurality of symbols, the symbol constellation extension projection having an outward angular region, the outward angular region defined by a value of an angle, and the value of the angle equal to or less than an angular distance between two symbols in the constellation. The value of the angle is based on a constellation and a code rate used for encoding and transmitting the stream of data.

Abstract: An information processing apparatus includes an estimation unit configured to estimate, based on a feature of a person of interest contained in an image captured in each of a plurality of regions and a feature of a previously set registered person, a certainty that the person of interest is the registered person, and an updating unit configured to, with respect to a first person of interest subjected to image capturing in a first region out of the plurality of regions, in a case where a period for which a state in which a certainty that a second person of interest contained in an image captured in a second region different from the first region is the registered person is larger than a threshold value is kept is longer than a predetermined period, perform updating in such a way as to lower a certainty that the first person of interest in the first region is the registered person, and, in a case where the period for which the state in which the certainty that the second person of interest is the registered per

Abstract: Methods and systems are provided for manufacturing an appliance for correcting malocclusions of a patient's teeth. The method may include measuring the positions of a patient's teeth and receiving tooth movement constraints. The method may also include generating an initial treatment plan based on the measured tooth positions and the tooth movement constraints and measuring the malocclusions of the patient's teeth for one or more types of dental malocclusions. The method may also include generating a plurality of treatment plans from the initial treatment plan based on the measured malocclusion and generating a model of an appliance for each stage of the treatment plan. The method may also include generating instructions for fabricating the appliance for each stage of the treatment plan based on the model of the appliance for each stage of the treatment plan.

Abstract: Embodiments of the present disclosure relate to a method and system to generate a waveform in a communication network. The transmitter receives an input data and transmit a generated waveform to another communication system. The input data is spread with a spread code to generate a spread data and rotated using a constellation rotation operation to produce a rotated data. The rotated data is then precoded using precoding filter to produce a precoded data, and transformed into DFT output data using DFT operation. The DFT output data is then mapped with subcarriers to generate the sub-carrier mapped DFT data and modulated using Orthogonal Frequency Division Multiplexing (OFDM) modulation to generate the waveform with low PAPR.

Abstract: The present disclosure provides a data processing method and a computing device. The data processing method is adapted to process an event data flow from a DVS to acquire image frames representing changes in a scene within different time periods. The event data flow includes coordinates of each pixel whose brightness has changed. The data processing method includes: segmenting the event data flow into a plurality of data slices at a predetermined time interval; and selecting a first quantity of data slices sequentially and performing weighted calculation on the first quantity of data slices to generate the image frames representing the changes in the scene within different time periods.

Abstract: A motion prediction system for predicting the motion of a random animate subject. A first neural network is a markerless motion capture network, trained to receive video data of the subject and to process the video data to generate a time sequence of musculoskeletal motion capture data. A second neural network is a motion prediction network, trained to receive the musculoskeletal motion capture data and to process the data to generate a prediction of the subject's location based on position change in position of joints and/or muscles.

Abstract: A receiver includes a continuous-time equalizer, a decision-feedback equalizer (DFE), data and error sampling logic, and an adaptation engine. The receiver corrects for inter-symbol interference (ISI) associated with the most recent data symbol (first post cursor ISI) by establishing appropriate equalization settings for the continuous-time equalizer based upon a measure of the first-post-cursor ISI.

Abstract: Systems and methods are described for image classification. An example method may comprise receiving an image comprising an object of interest and determining, based on a first convolution neural network, a first classification of the image. The first convolution neural network may be optimized for a first factor. The method may comprise determining, based on a second convolution neural network, a first classification of the image. The second convolution neural network may be optimized for a second factor. The method may comprise determining, based on the first classification and the second classification, a characteristic of the object of interest. The method may comprise providing the characteristic of the object of interest.