Abstract: A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.

Abstract: Methods and systems for detecting defects on a specimen are provided. One system includes one or more computer systems and one or more components executed by the one or more computer systems. The component(s) include a deep learning model configured for, for a location on a specimen, generating a gray scale simulated design data image from a high resolution image generated at the location by a high resolution imaging system. The computer system(s) are configured for generating a simulated binary design data image for the location from the gray scale simulated design data image. The computer system(s) are also configured for detecting defects at the location on the specimen by subtracting design data for the location from the simulated binary design data image.

Abstract: According to certain embodiments, an electronic device comprises a short-range communication module, a processor operatively connected to the short-range communication module, and a memory operatively connected to the processor, wherein the memory may store instructions which, when executed, cause the processor to: confirm that the electronic device is inserted into a first external electronic device; obtain setting information regarding a Bluetooth communication connection from the first external electronic device by using the short-range communication module; confirm whether a condition for setting address information of the electronic device is satisfied; and on the basis of confirming that the condition is satisfied, set the address information of the electronic device by using the setting information. Various other embodiments are possible.

Abstract: Techniques for training a machine learning model based on captured images are described. A method described include filtering a first set of collected images using one or more machine learning models; labeling the first set of filtered, collected images using a data labeling service using a service of the provider network; training a machine learning model from a machine learning algorithm using the first set of filtered, collected images using a service of the provider network; and causing deployment of the trained machine learning model onto a device.

Abstract: This document describes low-latency Bluetooth connectivity in a wireless network in which a central node and a peripheral node establish a connection. During a first connection interval, the peripheral node receives a packet from the central node to synchronize communication with the central node, and based on receiving the packet, the peripheral node transmits a first fixed-length packet. If the first fixed-length packet fails to reach the central node, the peripheral node does not receive an acknowledgement, ACK, from the central node during the first connection interval and retransmits the first fixed-length packet during the first connection interval.

Abstract: Aspects of the disclosure relate to monitoring devices at enterprise locations using machine-learning models to protect enterprise-managed information and resources. In some embodiments, a computing platform may receive, from one or more data source computer systems, passive monitoring data. Based on applying a machine-learning classification model to the passive monitoring data received from the one or more data source computer systems, the computing platform may determine to trigger a data capture process at an enterprise center. In response to determining to trigger the data capture process, the computing platform may initiate an active monitoring process to capture event data at the enterprise center. Thereafter, the computing platform may generate one or more alert messages based on the event data captured at the enterprise center. Then, the computing platform may send the one or more alert messages to one or more enterprise computer systems.

Abstract: Aspects of the disclosure relate to monitoring devices at enterprise locations using machine-learning models to protect enterprise-managed information and resources. In some embodiments, a computing platform may receive, from one or more data source computer systems, passive monitoring data. Based on applying a machine-learning classification model to the passive monitoring data received from the one or more data source computer systems, the computing platform may determine to trigger a data capture process at an enterprise center. In response to determining to trigger the data capture process, the computing platform may initiate an active monitoring process to capture event data at the enterprise center. Thereafter, the computing platform may generate one or more alert messages based on the event data captured at the enterprise center. Then, the computing platform may send the one or more alert messages to one or more enterprise computer systems.

Abstract: A training method for an anomaly detection model and an electronic device using the same are provided. The anomaly detection model includes a generative model and a discriminative model. The training method for the anomaly detection model includes the following steps. One of a plurality of original images and one of a plurality of task information are used as a training sample. The training sample is input to the generative model and the discriminative model to calculate a plurality of network loss results corresponding to the training sample. If the original image of the training sample does not match the task information of the training sample, a first loss function is obtained based on a weighted sum of reciprocals of the network loss results, and the generative model is trained according to the first loss function.

Abstract: Various embodiments are directed to systems and methods for providing employee data corresponding to individual employees to a customer computing entity operated by a customer. The employee data is provided so as to provide additional detail regarding an employee involved in a customer service interaction with the customer. The employee data is broadcast from an employee beacon carried by a respective employee. The employee data is received by a customer computing entity within the broadcast range of the employee beacon and, upon detection of a trigger event, the employee data is presented to the user via a graphical user interface of the customer computing entity, which requests that the customer provide feedback data regarding the employee, which may be transmitted to a management computing entity for storage and analysis.

Abstract: Embodiments provide a method for transmitting data from a user terminal to another apparatus. The method has a step of generating a signal for driving an electromagnetic actuator of a loudspeaker of the user terminal, and a step of driving the electromagnetic actuator by the generated signal to produce, by the electromagnetic actuator, a magnetic field which carries the data.

Abstract: A method includes sending, by a computing device, a first communication to a battery device having a system on a chip (SoC) that is powering another device, receiving, by the computing device, a second communication in response to the first communication from the battery device, the second communication comprising a received signal strength indicator (RSSI) from the battery device to the computing device, displaying, by the computing device, a graphical user interface (GUI) that indicates the RSSI that provides real-time information in response to at least one of movement of the computing device and movement of the battery device, and sending, by the computing device, a third communication to the battery that when received by the battery device causes the battery device to provide at least one audible alert.

Abstract: A medical care support device includes: an acquisition unit that acquires medical information including medical image data obtained by capturing a digestive tract of a subject; a derivation unit that derives presence or absence of a foreign object in the digestive tract, based on the medical information and a learned model learned in advance using plural pieces of learning medical information including the medical image data in which a label according to a kind of the foreign object is assigned to the foreign object in the digestive tract according to each organ of the digestive tract, and derives at least one of position, size, or kind of the foreign object if the foreign object is present; and an output unit that outputs removal information of the foreign object according to the at least one of position, size, or kind of the foreign object, based on a result of the derivation.

Abstract: Disclosed is a face verification method and apparatus. The method including analyzing a current frame of a verification image, determining a current frame state score of the verification image indicating whether the current frame is in a state predetermined as being appropriate for verification, determining whether the current frame state score satisfies a predetermined validity condition, and selectively, based on a result of the determining of whether the current frame state score satisfies the predetermined validity condition, extracting a feature from the current frame and performing verification by comparing a determined similarity between the extracted feature and a registered feature to a set verification threshold.

Abstract: Disclosed is a face verification method and apparatus. The method including analyzing a current frame of a verification image, determining a current frame state score of the verification image indicating whether the current frame is in a state predetermined as being appropriate for verification, determining whether the current frame state score satisfies a predetermined validity condition, and selectively, based on a result of the determining of whether the current frame state score satisfies the predetermined validity condition, extracting a feature from the current frame and performing verification by comparing a determined similarity between the extracted feature and a registered feature to a set verification threshold.

Abstract: Systems and methods are presented for establishing a communication link between two or more electronic devices. A portable eyewear electronic device is configured to communicate with a handheld electronic device, such as a ring, that in turn is retained by an accessory electronic device to establish a wired communication link. The accessory electronic device may be retained or housed by a second accessory electronic device, such as a remote control or wearable device.

Abstract: An electronic device is provided.

Abstract: A Bluetooth audio broadcasting system includes: an audio broadcasting device arranged to operably broadcast BLE audio packets; a first Bluetooth member device arranged to operably parse the BLE audio packets to acquire a predetermined audio data and to operably control a first audio playback circuit to playback the predetermined audio data; and a second Bluetooth member device arranged to operably parse the BLE audio packets to acquire the predetermined audio data and to operably control a second audio playback circuit to playback the predetermined audio data. After receiving the BLE audio packets from the audio broadcasting device, the first Bluetooth member further adjusts an audio volume of the first audio playback circuit according to a volume adjusting instruction transmitted from another device other than the audio broadcasting device, and the second Bluetooth member further adjusts an audio volume of the second audio playback circuit according to the volume adjusting instruction.

Abstract: A machine-learning architecture may be trained to determine point cloud data associated with different types of sensors with an object detected in an image and/or generate a three-dimensional region of interest (ROI) associated with the object. In some examples, the point cloud data may be associated with sensors such as, for example, a lidar device, radar device, etc.

Abstract: A method and system implementing a method for detecting a catheter in fluoroscopic data and updating a displayed electromagnetic position of the catheter on a 3D rendering is provided including navigating a catheter to a target area and acquiring fluoroscopic data from a fluoroscopic sweep of the target area. An initial catheter detection is performed to detect catheter tip candidates in each 2D frame of the fluoroscopic data using a shallow neural network. A secondary catheter detection is performed to detect catheter tip candidates in each 2D frame of the fluoroscopic data using a deep neural network. False-positive catheter tip candidates are removed by reconstructing a 3D position of the catheter tip and finding an intersecting point of rays corresponding to each 2D frame.

Abstract: A GaN-on-Si device structure and a method of fabrication are disclosed for improved die yield and device reliability of high current/high voltage lateral GaN transistors. A plurality of conventional GaN device structures comprising GaN epi-layers are fabricated on a silicon substrate (GaN-on-Si die). After processing of on-chip interconnect layers, a trench structure is defined around each die, through the GaN epi-layers and into the silicon substrate. A trench cladding is provided on proximal sidewalls, comprising at least one of a passivation layer and a conductive metal layer. The trench cladding extends over exposed surfaces of the GaN epi-layers, over the interface region with the substrate, and also over the exposed surfaces of the interconnect layers. This structure reduces risk of propagation of dicing damage and defects or cracks in the GaN epi-layers into active device regions. A metal trench cladding acts as a barrier for electro-migration of mobile ions.