Abstract: Disclosed are techniques for lane estimation. In aspects, a method includes receiving a plurality of camera frames captured by a camera sensor of a vehicle, assigning a first subset of the plurality of camera frames to a deep learning (DL) detector and a second subset of the plurality of camera frames to a computer vision (CV) detector based on availability of the DL and CV detectors, identifying a first set of lane boundary lines in a first camera frame processed by the DL detector, identifying a second set of lane boundary lines in a second camera frame processed by the CV detector, generating first and second sets of lane models based on the first and second sets of lane boundary lines, and updating a set of previously identified lane models based on the first set of lane models and/or the second set of lane models.

Abstract: A data processing system may include a plurality of memory modules, a controller, a power supply and a plurality of switches. Each of the memory modules may include a plurality of pages. The controller may control operations of the memory modules. The power supply may provide the memory modules with a power. The switches may be arranged corresponding to each of the memory modules. The switches may be selectively driven in response to a switch drive signal from the controller.

Abstract: Provided is a hologram printing method and apparatus using a hologram medium light efficiency map. A hologram printing method according to an embodiment emits a laser to a hologram medium, acquires an image by photographing light diffracted from the hologram medium, generates a light efficiency map of the hologram medium from the acquired image, and records hogels on the hologram medium by referring to the generated light efficiency maps of the hologram medium. Accordingly, light efficiency is measured on each hogel area, and hologram printing is performed by adjusting an intensity of a laser of each wavelength according to a hogel, so that uniformity of luminance and color of a hologram printing result can be enhanced.

Abstract: A display device includes a substrate, a scan line disposed on the substrate, a storage line parallel to the scan line and a data line intersecting the scan line and the storage line. First and second transistors are electrically connected to the scan line and the data line. A first pixel electrode is disposed on the first transistor and includes a first stem portion, a first contact portion and a first branch portion. The first branch portion electrically connects the first contact portion to the first stem portion, a second pixel electrode is disposed on the second transistor and includes a second stem portion, a second contact portion and a second branch portion. The second branch portion electrically connects the second contact portion to the second stem portion. A first pattern is disposed between the substrate and the second pixel electrode. The first pattern overlaps the second branch portion.

Abstract: A display device includes a substrate, a first data line, a second data line and a third data line on the substrate, each of the first to third data lines being extend in a first direction and arranged in a second direction crossing the first direction, a first power line which is between the first data line and the second data line on the substrate, extends in the first direction, and supplies a first supply voltage, and a second power line which is between the second data line and the third data line on the substrate, extends in the first direction, and supplies a second supply voltage equal to the first supply voltage.

Abstract: A controller that controls a nonvolatile memory apparatus may include a first memory configured to temporarily store user data, a second memory including a plurality of memory regions composed of one or more meta regions for storing meta data and at least one spare region, and a processor configured to control the first memory and the second memory and perform first start-gap wear leveling on at least one meta region using the at least one spare region as a gap.

Abstract: A display device and a method for fabricating a display device are provided. The display device includes a display panel, a printed circuit film, and an adhesive member interposed between the display panel and the printed circuit film, wherein the display panel comprises an insulating layer and a plurality of grooves overlapping with the adhesive member in a thickness direction of the display device and disposed on a surface of the insulating layer, and wherein the plurality of grooves is filled with the adhesive member.

Abstract: Described herein is a process, comprising: a) selecting an API Group II base stock with selected viscosity index and pour point; b) blending a base oil with the base stock, and c) adding to the base oil: i) a liquid ethylene propylene copolymer viscosity modifier that reduces a traction coefficient, and ii) an additive package, to make a driveline fluid that has a defined viscosity index and excellent shear stability. Also provided is a driveline fluid composition having the high viscosity index and excellent shear stability, comprising: a) a base oil comprising from 50 wt % to 100 wt % API Group II base stock; b) a liquid ethylene propylene copolymer viscosity modifier that reduces a traction coefficient; and c) an additive package. Further provided is a method for lubricating an axle or manual transmission by supplying the driveline fluid composition.

Abstract: An Out-Of-Distribution (OOD) detection method performed by an OOD detection device is provided. The OOD detection method includes generating a noise filter using training data based on a target distribution and using a classification model pre-trained according to the target distribution, training the generated noise filter to reduce a loss function for a first output value obtained by applying the training data passed through the generated noise filter to the pre-trained classification model, and detecting Out-Of-Distribution (OOD) for new data using a second output value obtained by applying the new data passed through the trained noise filter to the pre-trained classification model.

Abstract: A local model training method of a federated learning framework implementing training data classification is provided. In the local model training method, a client may classify training data into two categories, generate a learning mini-batch by adjusting a ratio between samples classified into the two categories and included in the mini-batch to a preset ratio, and train a learning model using the mini-batch with the adjusted sample ratio.

Abstract: The present invention relates to a power-assisted liposuction device. The power-assisted liposuction device includes: a handpiece; a motor mounted inside the handpiece and connected with a power source to provide driving force; a reverse cam of which one end is connected to the motor; and a cannula connected to the other end of the reverse cam. The reverse cam converts a rotary motion of the motor into a linear reciprocating motion so that the cannula performs a reciprocating motion when the motor is operated.

Abstract: A data processing apparatus may include a master device, a slave device, and a controller configured to arbitrate communication between the master device and the slave device by: setting a respective Time-out Counter (TC) for each of requests transmitted from the master device, allocating one or more virtual channels to each of one or more request groups, the one or more virtual channels respectively corresponding to priority levels, associating a request with a virtual channel corresponding to the priority level of the request, for each request group, selecting one of the leading requests of the respective virtual channels according to the TCs and transmitting the selected request to the slave device.

Abstract: An electronic device is provided. The electronic device includes a camera, a communication interface, and a processor configured to capture an image of a background of a display device through the camera, to divide the captured image into a plurality of blocks, to compare a target gradation value with gradation values of the each of the plurality of blocks, to adjust the gradation values of the each of the plurality of blocks, and to transmit an image of which the gradation values are adjusted, to the display device through the communication interface, wherein the sizes of the plurality of blocks are identified based on a background pattern of the display device.

Abstract: A data processing system may include a plurality of memory modules, a controller, a power supply and a plurality of switches. Each of the memory modules may include a plurality of pages. The controller may control operations of the memory modules. The power supply may provide the memory modules with a power. The switches may be arranged corresponding to each of the memory modules. The switches may be selectively driven in response to a switch drive signal from the controller.

Abstract: The present invention relates to a neovascular-targeting contrast medium composition and a method for preparing same. The neovascular-targeting contrast medium composition according to the present invention exhibits high binding force to neovascularization-associated ?v?3 integrin, excellent tissue permeability and biostability, enables simple measurement in vitro, in vivo, or ex vivo, and thus is effective in the detection of neovascularization and in diagnosing diseases associated therewith, and therefore may be usefully employed in the relevant industries.

Abstract: A cementless type dental implant includes: an abutment having vertical slits spaced apart from each other around an upper portion thereof, and having elastic coupling protrusions disposed between the vertical slits and radially providing elasticity; a crown cap having a coupling portion formed at a lower portion therein so that the upper portion of the abutment is inserted therein, having a screw hole formed at an upper portion therein, having a coupling step formed at a center therein to couple the elastic coupling protrusions of the abutment, and having a crown integrally formed outside; and an abutment screw inserted and fastened in a screw insertion hole of the abutment and having a protrusion pressing portion formed at an upper end thereof and pushing the elastic coupling protrusions radially outward when being fastened such that the elastic coupling protrusions are coupled inside the coupling step of the crown cap.

Abstract: Provided is a method of removing, by a server, an abnormal client in federated learning. A method of removing, by a server, an abnormal client in federated learning may include receiving, from a user equipment (UE), first weight values trained in a first local model, generating a first client model based on the first weight values, validating the first client model by using a validation data set in order to determine whether the first client model is legitimate, and removing the first weight values based on the first client model not being legitimate.

Abstract: The present disclosure relates to a composition for a radar transmissive cover for a vehicle, which may enhance dielectric properties while excellently maintaining mechanical properties. The composition, as the composition forming the radar transmissive cover for the vehicle through which a radar beam radiated from a radar is transmitted, has a main material in which 70 to 90 wt % of Polybutylene Terephthalate (PBT) and 10 to 30 wt % of Polycarbonate (PC) are mixed and an additive containing at least reinforcing filler mixed.

Abstract: A display device and a manufacturing method thereof, and a display device including: a substrate including a display area, an opening area positioned in the display area, and a peripheral area positioned between the opening area and the display area; a transistor positioned in the display area of the substrate; a first electrode connected to the transistor; a second electrode positioned on the first electrode; an inorganic pattern portion positioned in the peripheral area of the substrate; and an intermediate layer positioned between the first electrode and the second electrode, and including at least one organic layer positioned on the inorganic pattern portion. The inorganic pattern portion includes a plurality of inorganic patterns spaced apart from each other.

Abstract: Disclosed are an ?v?3 integrin targeting single-domain antibody and various applications thereof. The ?v?3 integrin targeting single-domain antibody exhibits high binding ability to ?v?3 integrin related to angiogenesis, excellent tissue permeability, and biostability compared to conventional antibodies. Further, the single-domain antibody may be combined with fluorescent particles and thus may be easily measured in vitro, in vivo or ex vivo, and may be effective in detecting angiogenesis and diagnosing angiogenesis related diseases, therefore it may be usefully used in related industries.