Abstract: Provided herein are methods for treating macular degeneration, age-related macular degeneration, and photoreceptor degeneration in a subject including delivering to the subject electrical stimulation in the form of a ramp waveform, wherein the electrical stimulation is delivered through the skin around the eye. The methods provide for electrical stimulation at least most days per week.

Abstract: A hardly graphitizable carbonaceous material can be used in a negative electrode material for nonaqueous electrolyte secondary batteries (for example, a lithium ion battery) having not only high charge capacity but also high charge-discharge efficiency and being fully charged to be used and a method for producing the same. Such hardly graphitizable carbonaceous material may have an oxygen element content of 0.25% by mass or less.

Abstract: An ultra-high voltage direct current (DS) power cable may be provided via a conductor formed by twisting a plurality of strands, an inner semiconducting layer surrounding the conductor, an insulating layer surrounding the inner semiconducting layer; and an outer semiconducting layer surrounding the insulating layer, wherein the insulating layer is formed of an insulation composition containing a polyolefin resin and a crosslinking agent, and the insulating layer is divided into three equal parts in a thickness direction, the three equal parts including an inner layer, an intermediate layer and an outer layer.

Abstract: A device may include a processor. The processor may be configured to: receive, from one or more network components, key performance indicators (KPIs) and parameters that are associated with a User Equipment device; select a scheduling strategy, for data communications over a wireless link between the device and the UE, based on the KPIs and the parameters; apply the selected scheduling strategy to schedule data for transmission to the UE; and transmit the data to the UE based on the scheduling.

Abstract: In a fine dust monitoring method, it is checked whether or not a preset first measurement event occurs in an area covered by the monitoring device, an occurrence point of the first measurement event is photographed when the first measurement event occurs, and a fine dust concentration of a point where the first measurement event occurred by inputting a photographed video image according to the first measurement event is outputted to a pre-trained first deep learning model.

Abstract: A fine dust measurement system includes a video photographing device that photographs a target place, generates a converted image or converted data by converting all or a portion of the photographed image and transmits the converted image or converted data, and a fine dust measurement that receives the converted image or converted data from the video photographing device, and trains a deep learning model to output a fine dust concentration of the target place by inputting the received converted image or converted data into the deep learning model.

Abstract: An environment monitoring device includes a first case, a second case coupled to the first case, and a smart phone provided fixed in an inner space between the first case and the second case, wherein the smart phone is configured to measure a fine dust concentration around the environment monitoring device based on an image photographed with a camera, and display information on the measured fine dust concentration via a screen of the smart phone.

Abstract: A thermoelectric array according to one embodiment of the present invention comprises: a first thermoelectric module; and a second thermoelectric module arranged to be adjacent to the first thermoelectric module, wherein: the first thermoelectric module includes a first lower substrate, a plurality of semiconductor devices arranged on the first lower substrate, and a first upper substrate arranged on the plurality of semiconductor elements; the second thermoelectric module includes a second lower substrate, a plurality of semiconductor elements arranged on the second lower substrate, and a second upper substrate arranged on the plurality of semiconductor elements; and the first thermoelectric module includes a first groove, and the second thermoelectric module includes a second groove arranged at a position corresponding to the first groove.

Abstract: A brake body according to the present disclosure has a coating layer which contains an iron-based amorphous alloy and is formed on a contact surface at which a friction means rubs against the brake body to generate braking force which the brake body is rotated. The coating layer has a low friction coefficient and thus generates less dust during braking. The coating layer also has high corrosion resistance and wear resistance and may thus guarantee high performance and good price competitiveness when applied to a brake base material produced at low cost.

Abstract: There is provided a frame up-sampling-based key point smoothing method. The key point smoothing method according to an embodiment up-samples frames based on which key points are extracted, and smooths the up-sampled frames. Accordingly, key points are generated by smoothing key point extraction frames after up-sampling, so that a time-series error occurring when key points are extracted can be reduced and time-series stability can be enhanced, and quality of an application service provided subsequently can be improved.

Abstract: There is provided a streaming speech synthesis method and system for supporting a real-time conversation model. A real-time speech synthesis method according to an embodiment outputs a sentence for responding to an utterance of a user in the unit of a text through a conversation model, and synthesizes a speech in the unit of the outputted text through a speech synthesis model. Accordingly, a text which is a shorter unit than a sentence is continuously received and is immediately synthesized into a speech, so that speech synthesis can be performed in real time according to a speed of a real-time conversation model without delay even when a conversation is generated in the unit of a text in the real-time conversation model.

Abstract: The present disclosure relates to a golf club head that can increase strength and repulsion force, and includes a face portion formed in the front of the golf club head and directly contacting a golf ball to strike the golf ball, a crown portion extended rearward from the face portion to form an upper portion and a rear portion of the golf club head, a sole portion extended forward from the crown portion to form a bottom portion, a neck portion formed in an upper portion of the face portion and coupled with a shaft having a grip gripped by a user in the upper portion, and a reinforcing portion formed on a rear surface of the face portion, in which the reinforcing portion is formed to have an area narrower than that of the face portion on the rear surface of the face portion.

Abstract: A device may include a processor configured to receive a data stream connection request via a WIFI connection or a wired connection and establish a cellular wireless connection with a base station based on the data stream connection request. The processor may be further configured to determine a cellular wireless Quality of Service (QOS) class or network slice assigned to the cellular wireless connection by a core network associated with the base station; map the cellular wireless QoS class or network slice to a priority class associated with the WIFI connection or the wired connection; assign the priority class to a data stream associated with the data stream connection request; and process data units associated with the data stream based on the assigned priority class.

Abstract: Proposed is an automatic supply and assembly device for fasteners, the automatic supply and assembly device being capable of automatically aligning and supplying fasteners and assembling the fasteners. An automatic supply and assembly device for fasteners according to an embodiment includes a main frame, a rail having a guide groove guiding the fasteners, a storage box provided at a second end of the rail and storing the fasteners, a supply unit lifted in the storage box and supplying the fasteners stored in the storage box to the guide groove located at the second end of the rail, a holding plate located apart from a first end of the rail, a withdrawal unit individually withdrawing the fasteners received in the guide groove located at the first end of the rail to the holding plate, and an assembly driver movable upward and downward and assembling a fastener located in the holding plate.

Abstract: Proposed is an assembly system for a crash pad lower assembly panel, the assembly system being capable of automatically aligning and supplying fasteners and assembling accessories to a crash pad lower assembly panel using the fasteners. An assembly system for a crash pad lower assembly panel according to an embodiment includes a work table, a jig provided on a first side of the upper surface of the work table and configured to support a front surface of the crash pad lower assembly panel, a pressing rod provided on the work table and configured to pressurize the crash pad lower assembly panel supported on the jig, an orthogonal robot provided on the work table and configured to be movable in the X-axis and Y-axis above the jig, and an assembly device coupled to the orthogonal robot.

Abstract: A thermoelectric device according to an embodiment of the present invention comprises: a first substrate; an insulating layer disposed on the first substrate; a first electrode unit disposed on the insulating layer; a first terminal electrode and a second terminal electrode disposed on the insulating layer and protruding from the first electrode unit toward the first outer side of the first substrate; a semiconductor structure disposed on the first electrode unit; a second electrode unit disposed on the semiconductor structure; and a second substrate unit disposed on the second electrode unit. The second substrate unit comprises a plurality of second substrates disposed away from one another. The first electrode unit comprises: a plurality of electrode groups respectively overlapping the plurality of second substrates vertically; and a first connection electrode connecting two different electrode groups among the plurality of electrode groups.

Abstract: In some implementations, a network element may identify gaming traffic associated with a user equipment (UE). The network element may detect a traffic pattern associated with the gaming traffic. The network element may determine that the UE is associated with a prioritized service, wherein the prioritized service is associated with network slicing. The network element may perform a network scheduling for the UE that prioritizes the gaming traffic associated with the UE over non-gaming traffic associated with another UE, wherein the network scheduling is based on the traffic pattern and the UE being associated with the prioritized service.

Abstract: There is provided a training method of a multi-task integrated deep learning model. A multi-task integrated deep learning model training method according to an embodiment may generate training data for a plurality of visual intelligence tasks from visual data in a batch, and may train a multi-task integrated deep learning model which performs a plurality of visual intelligence tasks by using the generated training data. Accordingly, training data for training an integrated deep learning model which performs various visual intelligence tasks is generated in a batch through multi-data conversion kernels, so that appropriate training data for performing multiple tasks may be easily obtained and effective training of a multi-task integrated deep learning model is possible.

Abstract: There is provided a training dataset construction method for speech synthesis through fusion of language, speaker, emotion within an utterance. A training dataset construction method of a speech synthesis model according to an embodiment collects speech data having different speech utterance information, increases the speech data by fusing the collected speech data within one utterance, and generates a training dataset by using the increased speech data. Accordingly, a training dataset for speech synthesis is constructed through fusion of language, speaker, emotion within one utterance, so that quality of speech synthesis of multi-speaker/multi-language/emotion can be enhanced.

Abstract: There is provided a speech synthesis system and method with an adjustable utterance length. The speech synthesis method according to an embodiment predicts a duration of each phoneme corresponding to a speech mask from the speech mask and a text to be synthesized with the speech mask, encodes the text to be synthesized and extracts a text sequence which is expressed by feature information of the text, generates a speech frame sequence by regulating a length of each phoneme of the text sequence according to the predicted duration of each phoneme corresponding to the speech mask, and synthesizes a speech from the generated speech frame sequence. Accordingly, a length of a speech to be synthesized can be freely regulated as a user desires by regulating a length of a speech mask.