Abstract: A semiconductor device includes a gate trench formed in a substrate, a gate dielectric layer formed along profile of sidewalls and a bottom surface of the gate trench, first and second gate electrodes that are stacked over the gate dielectric layer to gap-fill a portion of the gate trench, a dipole inducing portion positioned between the second gate electrode and the gate dielectric layer and including a dipole bond and a non-dipole bond, and a capping layer suitable for gap-filling a remaining portion of the gate trench over the dipole inducing portion and the second gate electrode.

Abstract: There is provided a method for dividing query computations and scheduling for CSDs in a DB system in which a plurality of CSDs are used as a storage. A scheduling method according to an embodiment includes: selecting one of a plurality of scheduling polices; selecting a CSD to which snippets included in a group are delivered according to the selected scheduling policy; and delivering the snippets to the selected CSD, and the scheduling polices are polices for selecting CSDs to which snippets are delivered, based on different criteria. Accordingly, CSDs may be randomly selected according to user setting or a query execution environment, or an optimal CSD may be selected according to a CSD status or a content of an offload snippet, so that a query execution speed can be enhanced.

Abstract: An image encoding/decoding method and apparatus are provided. An image decoding method includes obtaining inter prediction information of a current block and wraparound information from a bitstream, and generating a prediction block of the current block based on the inter prediction information and the wraparound information. The wraparound information may include a first flag specifying whether wraparound motion compensation is enabled for a current video sequence including the current block. The first flag may have a first value specifying that the wraparound motion compensation is disabled, based on that one or more subpicture, which is coded independently and has a width different from a width of a current picture, is present in the current video sequence.

Abstract: There is provided a smart power management method for power consumption reduction based on an intelligent BMC. A cooling fan control method by a BMC according to an embodiment includes: collecting monitoring data regarding computing modules; calculating a current CPU power from the collected monitoring data; predicting a future CPU temperature from the collected monitoring data; setting a rotation speed of a cooling fan based on the calculated current CPU power and the predicted future CPU temperature; and controlling the cooling fan at the set rotation speed. Accordingly, the BMC controls a cooling fan effectively/efficiently by interworking with on-device AI, thereby reducing power consumption in a server.

Abstract: A method of generating a human model according to the present disclosure may include generating a plurality of feature maps from an input image, wherein the plurality of feature maps include a body center map, a part index map, a body part map and a parameter map, generating a part-attentive feature configured with feature maps for each body part based on the part index map and the body part map, readjusting the part-attentive feature based on the parameter map and generating a pose parameter based on the readjusted part-attentive feature.

Abstract: The present embodiment relates to a dual lens drive device that comprises: a housing; a first bobbin which is disposed to move in a first direction inside the housing; a second bobbin which is disposed to move in the first direction inside housing and is spaced apart from the first bobbin; a first coil which is disposed on the first bobbin; a second coil which is disposed on the second bobbin; a magnet which is disposed in the housing and faces the first coil and the second coil; a base which is disposed below the housing; a substrate which comprises a circuit member having a third coil disposed to face the magnet between the housing and the base; and a support member which movably supports the housing with respect to the substrate, wherein the housing is integrally formed.

Abstract: A display device includes PAM data lines receiving PAM and PWM data voltages, and sub-pixels connected to the PAM and PWM data lines. A sub-pixel includes a light emitting element, a first pixel driver to supply a control current according to one of the PAM data voltages to a node, a second pixel driver to generate a driving current according to one of the PWM data voltages, and a third pixel driver to adjust a period during which the driving current is supplied to the light emitting element according to a voltage of the node. A peak current value of the driving current when the sub-pixel emits a light corresponding to a low gray level region is smaller than a peak current value of the driving current when the sub-pixel emits a light corresponding to a high gray level region higher than the low gray level region.

Abstract: A home port for a semiconductor manufacturing nozzle head includes a body having a discharge space configured to receive treatment liquid discharged from a plurality of nozzles, discharge flow passages connected to be in communication with the discharge space and penetrating through the body so as to face the plurality of nozzles, and a cleaning liquid distribution system, which is formed to penetrate through the body, and is connected to transfer a cleaning liquid to the discharge flow passages. The cleaning liquid distribution system includes supply flow passages connected to supply the cleaning liquid to the discharge flow passages, and a lead-in passage connected to and which joins the supply flow passages, and connected to receive the cleaning liquid injected from the outside.

Abstract: There is provided an intelligent BMC for predicting a fault by interworking on-device AI. A fault prediction method of a BMC according to an embodiment includes: collecting monitoring information regarding computing modules installed on a main board; calculating a FOFL from the collected monitoring data; and constructing an AI model related to the calculated FOFL and predicting a FOFL from the monitoring data. Accordingly, a fault occurring in various patterns may be predicted based on monitoring data by interworking with on-device AI.

Abstract: An aerosol-generating article according to one aspect of the present disclosure may include a tobacco medium portion, a filter portion disposed to be spaced apart from the tobacco medium portion, and a tubular cooling portion disposed between the tobacco medium portion and the filter portion. An inner peripheral surface of the cooling portion may include a nicotine component.

Abstract: There are a method and an apparatus for managing a hybrid cloud to perform consistent resource management for all resources in a heterogeneous cluster environment which is comprised of an on-premise cloud and a plurality of public clouds. Accordingly, the method and apparatus for hybrid cloud management provides an integration support function between different cluster orchestrations in a heterogenous cluster environment which is comprised of an on-premise cloud and a plurality of public clouds, supports consistent resource management for all resources, and provides optimal workload deployment, free optimal reconfiguration, migration and restoration, whole resource integration scaling.

Abstract: There is provided a cloud management method and apparatus for available GPU resource scheduling in a large-scale container platform environment. Accordingly, a list of available GPUs may be reflected through a GPU resource metric collected in a large-scale container driving (operating) environment, and an allocable GPU may be selected from the GPU list according to a request, so that GPU resources can be allocated flexibly in response to a GPU resource request of a user (resource allocation reflecting requested resources rather than 1:1 allocation).

Abstract: There is provided a query execution method in a DB system in which a plurality of CSDs are used as a storage. According to an embodiment, a query execution method includes: generating snippets for offloading a part of query computations for a query received from a client to CSDs; scheduling the generated snippets for the CSDs; collecting results of offloading; and merging the collected results of offloading. Accordingly, by dividing query computations, offloading, and processing in parallel, while processing query computations that are inappropriate for offloading by a DBMS, a query request from a client can be executed effectively and rapidly.

Abstract: A raw material feed hopper according to the present embodiment may include a hollow tube having a raw material accommodating space formed therein, a supporter supporting the hollow tube, a cone accommodated in the raw material accommodating space so as to be able to lift up and down, a rod connected to the cone, a connector connected to an upper portion of the rod, a lifting rod connected to the connector, and an elevator installed on the supporter for lifting up and down the lifting rod.

Abstract: Disclosed are polycyclic aromatic compounds that can be employed in various organic layers of organic electroluminescent devices. Also disclosed are organic electroluminescent devices including the polycyclic aromatic compounds. The organic electroluminescent devices are highly efficient and long lasting and have greatly improved luminous efficiency.

Abstract: The present technology may include an error correction code engine configured to generate a parity bit and syndrome information by performing an operation according to operation source data, and a data processing circuit configured to simultaneously output the parity bit and first delay data, which is generated by delaying input data by a first time according to a write operation, simultaneously output the syndrome information and second delay data, which is generated by delaying input data by a second time according to a read operation, and to share substantially the same signal path in generating the first delay data and in generating the second delay data.

Abstract: A touch sensor may include a substrate and may include electrode units, first demultiplexers, second demultiplexers, and driving pads all located on the substrate. The electrode units each may include a plurality of electrode groups, the electrode groups each including a plurality of touch electrodes. The first demultiplexers each may include a plurality of sub-demultiplexers and each may be electrically connected to a corresponding one of the electrode units. Each of the sub-demultiplexers of a first demultiplexer may be electrically connected to a corresponding one of the electrode groups of a corresponding electrode unit. The second demultiplexers may be connected between the first demultiplexers and the driving pads.

Abstract: The present application relates to a method for synthesizing image and an image synthesizing apparatus using the same. The image synthesizing method of the present application and the image synthesizing apparatus using the same can reduce the time required for learning by reducing the distortion phenomenon of the algorithm and reducing the amount of calculation required for image synthesis. In addition, the image synthesizing method of the present application and the image synthesizing apparatus using the same can be utilized in various fields such as national defense, IT, and entertainment based on characteristics such as excellent performance and learning time reduction of deep learning algorithms, and can be utilized in psychological warfare or induction of command system confusion.

Abstract: An aerosol generating system includes: a cavity configured to accommodate at least a portion of a cigarette; a first induction coil located around the cavity; a second induction coil located around the cavity and connected to the first induction coil in parallel; and a battery configured to supply an alternating current to the first induction coil and the second induction coil, wherein the first induction coil and the second induction coil have different resonant frequencies.

Abstract: A circularly polarizing plate capable of improving the tone of reflective color sense by using a retardation film having flat dispersion characteristics; and an OLED device comprising the circularly polarizing plate.