Abstract: The present invention improves a heat dissipation property of a semiconductor device by transferring hexagonal boron nitride (hBN) with a two-dimensional nanostructure to the semiconductor device. A semiconductor device of the present invention includes a substrate having a first surface and a second surface, a semiconductor layer formed on the first surface of the substrate, an hBN layer formed on at least one surface of the first surface and the second surface of the substrate, and a heat sink positioned on the second surface of the substrate. A radiation rate of heat generated during driving of an element is increased to decrease a reduction in lifetime of a semiconductor device due to a temperature increase. The semiconductor device has a structure and configuration which are very effective in improving a rapid temperature increase due to heat generated by high-power semiconductor devices.

Abstract: A direct cooling type semiconductor package unit includes a substrate made of a material capable of manufacturing a semiconductor device, and having a material layer for forming the semiconductor device stacked on one side of the substrate, and a flow channel through which a cooling fluid flows formed on the other side of the substrate to enable direct cooling of the semiconductor device using the cooling fluid; a packaging block disposed at a position spaced apart from the substrate for packaging the semiconductor device, and having an electrode electrically connected to the semiconductor device through wiring and placed thereon to be insulated; a heat sink unit disposed on a lower side of the packaging block and having a fluid movement region formed at a position corresponding to a flow channel of the substrate; and a thin film type structure disposed between the substrate and the heat sink unit for coupling between the substrate and the heat sink unit and being moldable to have pattern structures of variou

Abstract: Provided is a direct cooling device for an integrated circuit configured to form a direct cooling portion with a flow channel through which cooling fluid may flow in a through via hole of a substrate constituting the integrated circuit, and to couple the substrate to a heat sink unit through a bonding portion formed integrally with the direct cooling portion, unlike the prior art in which only a ground circuit is possible through the through via hole, which derives the effect of increasing product reliability due to improved thermal management efficiency of the integrated circuit by directly cooling the semiconductor device as well as the ground circuit, and the effect of simplifying and miniaturizing the structure by implementing the cooling function using a circuit for grounding the semiconductor device even without forming an additional flow path structure for cooling the semiconductor device.

Abstract: Disclosed herein are a learning-based three-dimensional (3D) model creation apparatus and method. A method for operating a learning-based 3D model creation apparatus includes generating multi-view feature images using supervised learning, creating a three-dimensional (3D) mesh model using a point cloud corresponding to the multi-view feature images and a feature image representing internal shape information, generating a texture map by projecting the 3D mesh model into three viewpoint images that are input, and creating a 3D model using the texture map.

Abstract: A method for realizing a real-time spatial structure in an apparatus for realizing a real-time spatial structure is provided. The method for realizing a real-time spatial structure includes: tracking a movement path of a player; and realizing a spatial structure in the space on the movement path of the player using at least one of water sprayed through water spray modules of a mesh structure installed on a ceiling and electromagnet ropes of the mesh structure.

Abstract: Disclosed herein are an apparatus and method for guiding multi-view capture. The apparatus for guiding multi-view capture includes one or more processors and an execution memory for storing at least one program that is executed by the one or more processors, wherein the at least one program is configured to receive a single-view two-dimensional (2D) image obtained by capturing an image of an object of interest through a camera, generate an orthographic projection image and a perspective projection image for the object of interest from the single-view 2D image using an image conversion parameter that is previously learned from multi-view 2D images for the object of interest, generate a 3D silhouette model for the object of interest using the orthographic projection image and the perspective projection image, and output the 3D silhouette model and a guidance interface for the 3D silhouette model.

Abstract: Disclosed herein are a learning-based three-dimensional (3D) model creation apparatus and method. A method for operating a learning-based 3D model creation apparatus includes generating multi-view feature images using supervised learning, creating a three-dimensional (3D) mesh model using a point cloud corresponding to the multi-view feature images and a feature image representing internal shape information, generating a texture map by projecting the 3D mesh model into three viewpoint images that are input, and creating a 3D model using the texture map.

Abstract: Disclosed herein are a learning-based three-dimensional (3D) model creation apparatus and method. A method for operating a learning-based 3D model creation apparatus includes generating multi-view feature images using supervised learning, creating a three-dimensional (3D) mesh model using a point cloud corresponding to the multi-view feature images and a feature image representing internal shape information, generating a texture map by projecting the 3D mesh model into three viewpoint images that are input, and creating a 3D model using the texture map.

Abstract: A method and a system for generating a 3D image of a character through steps of: receiving an identifier of a template and a two-dimensional (2D) image of the character generated based on the template corresponding to a type of the character; acquiring template information of the template from a template library by using the identifier, and extracting an effective parameter for machine learning to be performed to generate the 3D image based on the template information and the 2D image; and generating the 3D image of the character by performing the machine learning based on the effective parameter and the template information are provided.

Abstract: Disclosed herein are an apparatus and method for generating a 3D model. The apparatus for generating a 3D model includes one or more processors, and an execution memory for storing at least one program that is executed by the one or more processors, wherein the at least one program is configured to receive two-dimensional (2D) original image layers for respective viewpoints, and generate pieces of 2D original image information for respective objects by performing original image alignment on the 2D original image layers for respective viewpoints for each predefined object type, generate 3D model layers for respective objects from the pieces of 2D original image information for respective objects using multiple learning models corresponding to the predefined object types, and generate a 3D model by synthesizing the 3D model layers for respective objects.

Abstract: Disclosed herein are an apparatus and method for guiding multi-view capture. The apparatus for guiding multi-view capture includes one or more processors and an execution memory for storing at least one program that is executed by the one or more processors, wherein the at least one program is configured to receive a single-view two-dimensional (2D) image obtained by capturing an image of an object of interest through a camera, generate an orthographic projection image and a perspective projection image for the object of interest from the single-view 2D image using an image conversion parameter that is previously learned from multi-view 2D images for the object of interest, generate a 3D silhouette model for the object of interest using the orthographic projection image and the perspective projection image, and output the 3D silhouette model and a guidance interface for the 3D silhouette model.

Abstract: Provided are an ethylene-based polymer having excellent long-term pressure resistance characteristics and a pipe using the ethylene-based polymer. The ethylene-based polymer satisfies the balance of mechanical characteristics and excellent molding processability, as compared with a conventional ethylene-based polymer. Also provided are an ethylene-based polymer having a wide molecular weight distribution and a small lamellar thickness, thereby increasing tie molecules and obtaining excellent long-term pressure resistance characteristics, and a pipe using the same.

Abstract: A camera auxiliary device for privacy protection and a privacy protection method using the camera auxiliary device. The camera auxiliary device for privacy protection includes a processor for splitting an input light beam that is reflected from a capturing target into a first input beam for detecting a privacy protection area and a second input beam to be transferred to a camera connected to a user terminal, detecting a privacy protection area in an image signal generated based on the first input beam, and converting the second input beam and then transferring a converted second input beam to the camera so that personal information included in the privacy protection area is not visually identified, and a memory for storing the image signal and the privacy protection area.

Abstract: Disclosed herein are a learning-based three-dimensional (3D) model creation apparatus and method. A method for operating a learning-based 3D model creation apparatus includes generating multi-view feature images using supervised learning, creating a three-dimensional (3D) mesh model using a point cloud corresponding to the multi-view feature images and a feature image representing internal shape information, generating a texture map by projecting the 3D mesh model into three viewpoint images that are input, and creating a 3D model using the texture map.

Abstract: A method and a system for generating a 3D image of a character through steps of: receiving an identifier of a template and a two-dimensional (2D) image of the character generated based on the template corresponding to a type of the character; acquiring template information of the template from a template library by using the identifier, and extracting an effective parameter for machine learning to be performed to generate the 3D image based on the template information and the 2D image; and generating the 3D image of the character by performing the machine learning based on the effective parameter and the template information are provided.

Abstract: Disclosed herein are an apparatus and method for generating a training data set for machine learning. The method for generating a training data set, performed by the apparatus for generating the training data set for machine learning, includes generating a 3D model for a deformed 3D character based on 3D data pertaining to the 3D character, generating a 2D image corresponding to the 3D model, and generating the training data set for machine learning, through which the 3D character is generated from the 2D image, using the 2D image and the 3D model.

Abstract: An apparatus for generating a 3D printing model using multiple textures includes a template model storage unit configured to store template models that are 3D models previously manufactured to be printable, a 3D model conversion unit configured to, upon receiving a 3D input model and a texture image, perform mesh deformation on the template model stored in the template model storage unit by using geometric information about the 3D input model, a texture processing unit configured to assign the texture image to the 3D model having being subjected to the mesh deformation, and a 3D printing model output unit configured to output a geometric model and a texture image of the template model which are finally calculated.

Abstract: Provided is a method and apparatus for reconstructing a three-dimensional (3D) face based on a stereo camera, the method including: acquiring n images of a target by controlling a plurality of stereo cameras in response to an image acquirement request, wherein n denotes a natural number; extracting n face regions from the n images, respectively; and reconstructing a viewpoint-based face image based on the n face regions.

Abstract: Disclosed herein is an apparatus and method for automatically creating a 3D personalized figure suitable for 3D printing by detecting a face area and features for respective regions from face data acquired by heterogeneous sensors and by optimizing global/local transformation. The 3D personalized figure creation apparatus acquires face data of a user corresponding to a reconstruction target; extracts feature points for respective regions from the face data, and reconstructs unique 3D models of the user's face, based on the extracted feature points; creates 3D figure models based on the unique 3D models and previously stored facial expression models and body/adornment models; and verifies whether each 3D figure model has a structure and a shape corresponding to actual 3D printing, corrects and edits the 3D figure model based on results of verification, and outputs a 3D figure model corresponding to 3D printing.

Abstract: An apparatus for generating a 3D printing model using multiple textures includes a template model storage unit configured to store template models that are 3D models previously manufactured to be printable, a 3D model conversion unit configured to, upon receiving a 3D input model and a texture image, perform mesh deformation on the template model stored in the template model storage unit by using geometric information about the 3D input model, a texture processing unit configured to assign the texture image to the 3D model having being subjected to the mesh deformation, and a 3D printing model output unit configured to output a geometric model and a texture image of the template model which are finally calculated.