Abstract: A semantic segmentation method performed by at least one computing device is provided. The method may include extracting a plurality of support features, which include a first-level feature and a second-level feature, from a support image; generating a plurality of prototypes for a class associated with the support image, based on the plurality of support features; and predicting a segmentation mask for a query image using the plurality of prototypes, wherein the plurality of prototypes comprise a first prototype generated based on the first-level feature and a second prototype generated based on the second-level feature.

Abstract: A method for comparing documents and a system therefor are provided. The method according to some embodiments may include acquiring a first document image and a second document image, extracting a first feature set from the first document image and a second feature set from the second document image through an encoder, generating a correlation feature set by analyzing a correlation between at least part of the first feature set and at least part of the second feature set, and outputting a result of comparison between the first document image and the second document image based on a result of decoding the correlation feature set.

Abstract: An organic compound adsorbent and a gas blowing apparatus comprising same, wherein the organic compound adsorbent has a BET specific surface area of 500 m2/g or more, and the relationship between an average diameter of unit pore inlets and an average diameter inside unit pores of the metal organic framework satisfies Equation 1: 0.4 ? x ? ( nm ) / y ? ( nm ) ? 1. [ Equation ? 1 ] wherein: x is an average diameter of a unit pore inlet of the metal organic framework; and y is an average diameter inside a unit pore of the metal organic framework, and thus even in a dynamic gas flow having a flow rate of 1 cm/sec or more, the adsorption amount and the adsorption rate of volatile organic compounds can be improved, and therefore volatile organic compounds can be efficiently adsorbed in environments having a gas flow, such as an air purifier and an air conditioner.

Abstract: The present invention relates to a fire resistant paint composition, to a production method for same and to a painting method for a fire resistant paint using same, and, one example of implementation of the present invention can provide a fire resistant paint composition comprising: between 70 and 95 wt. % of a binder; between 1 and 10 wt. % of an aerogel; between 1 and 5 wt. % of a foaming agent; and the remainder of water, and can provide a production method for same and a painting method for a fire resistant paint using same.

Abstract: A semantic segmentation method performed by at least one computing device is provided. The method may include extracting a plurality of support features, which include a first-level feature and a second-level feature, from a support image; generating a plurality of prototypes for a class associated with the support image, based on the plurality of support features; and predicting a segmentation mask for a query image using the plurality of prototypes, wherein the plurality of prototypes comprise a first prototype generated based on the first-level feature and a second prototype generated based on the second-level feature.

Abstract: A method for auto-generating an AutoCAD drawing includes providing an interface for extracting only input data required for drawing equipment from strength calculation data, displaying all components and nozzles constituting the equipment and providing or correcting information thereon. The AutoCAD drawing is automatically generated based on information on all components constituting the equipment and nozzles.

Abstract: In a conversation contents management method for a conversation contents management system including at least one processor and a memory, a conversation party, conversation contents, and a conversation time included in a conversation section determined by at least one of a start point and an end point selected by a user are identified. The conversation section including the identified conversation party, conversation contents, and conversation time are stored in the memory as a unit conversation.

Abstract: The present invention relates to a method for fabrication of an extracellular matrix-induced self-assembly and to fabrication of an artificial tissue using same. The method for fabrication of an extracellular matrix-induced self-assembly comprise the steps of: (a) decellularizing and powdering a tissue-derived extracellular matrix (ECM); and (b) adding the decellularized extracellular matrix powder to cells and culturing the cells to form a cell-extracellular matrix powder self-assembly. Accordingly, the self-assembly has characteristics similar to those of extracellular matrix tissues and can be fabricated into three-dimensional artificial tissues 1 cm or greater in size, thus finding advantageous applications as a cell therapy product and an artificial tissue implant.

Abstract: The present invention relates to a composition for a moisture permeation preventing film for an electronic device, a method for forming a moisture permeation preventing film for an electronic device using the same, and an electronic device. The present invention can provide: a composition for a moisture permeation preventing film for an electronic device, the composition containing 1-10 wt % of an aerogel, 10-50 wt % of a solvent, 0.1-5 wt % of an additive, and the remainder binder; a method for forming a moisture permeation preventing film for an electronic device using the same; and an electronic device.

Abstract: A method for auto-generating an AutoCAD drawing includes providing an interface for extracting only input data required for drawing equipment from strength calculation data, displaying all components and nozzles constituting the equipment and providing or correcting information thereon. The AutoCAD drawing is automatically generated based on information on all components constituting the equipment and nozzles.

Abstract: A method for training a deep learning model according to a disclosed embodiment is performed by a computing device comprising one or more processors and a memory in which one or more programs to be executed by the one or more processors are stored, and the method includes acquiring first training data and extracting training feature vectors from the first training data, classifying the first training data into a plurality of groups based on the extracted training feature vectors, extracting ground-truth feature vectors from ground truths labeled on the first training data, classifying the ground-truth feature vectors into a plurality of groups corresponding to groups of the training feature vectors, calculating group reference information for each group of the ground-truth feature vectors, and setting a quality weight for second training data using the group reference information for each group.

Abstract: Disclosed is a device for supporting hoop stress applied to a meniscus and preventing deviation of the meniscus, and the device for supporting hoop stress applied to a meniscus and preventing deviation of the meniscus may include an intermediate support member, one side extension member, and the other side extension member.

Abstract: The present invention relates to a composition which is for preventing and treating eye diseases and comprises an anti-angiopoietin-2 (Ang2) antibody and, more specifically, to a novel use of an anti-Ang2 antibody as an agent for preventing or treating eye diseases, wherein the anti-Ang2 antibody is specifically coupled to Ang2, is coupled to a Tie2 receptor together with Ang2, and has improved affinity. The composition according to the present invention can be usefully used to develop a treatment agent for macular degeneration, diabetic retinopathy, and glaucoma.

Abstract: The present invention relates to a composition which is for preventing and treating eye diseases and comprises an anti-angiopoietin-2 (Ang2) antibody and, more specifically, to a novel use of an anti-Ang2 antibody as an agent for preventing or treating eye diseases, wherein the anti-Ang2 antibody is specifically coupled to Ang2, is coupled to a Tie2 receptor together with Ang2, and has improved affinity. The composition according to the present invention can be usefully used to develop a treatment agent for macular degeneration, diabetic retinopathy, and glaucoma.

Abstract: Disclosed is a device for supporting hoop stress applied to a meniscus and preventing deviation of the meniscus, and the device for supporting hoop stress applied to a meniscus and preventing deviation of the meniscus may include an intermediate support member, one side extension member, and the other side extension member.

Abstract: A method of providing a chat service between a first device and a plurality of second devices, includes: creating a first chat room configured to display visual signs transmitted and received between the first device and the plurality of second devices; receiving a command to select one or more of the plurality of second devices while the first chat room is displayed on the first device; selecting the one or more of the plurality of second devices based on the command; and creating a second chat room configured to display visual signs transmitted and received between the first device and the selected one or more of the plurality of second devices.

Abstract: The present invention relates to a composition for a moisture permeation preventing film for an electronic device, a method for forming a moisture permeation preventing film for an electronic device using the same, and an electronic device. The present invention can provide: a composition for a moisture permeation preventing film for an electronic device, the composition containing 1-10 wt % of an aerogel, 10-50 wt % of a solvent, 0.1-5 wt % of an additive, and the remainder binder; a method for forming a moisture permeation preventing film for an electronic device using the same; and an electronic device.

Abstract: The present invention relates to a fire resistant paint composition, to a production method for same and to a painting method for a fire resistant paint using same, and, one example of implementation of the present invention can provide a fire resistant paint composition comprising: between 70 and 95 wt. % of a binder; between 1 and 10 wt. % of an aerogel; between 1 and 5 wt. % of a foaming agent; and the remainder of water, and can provide a production method for same and a painting method for a fire resistant paint using same.

Abstract: In a conversation contents management method for a conversation contents management system including at least one processor and a memory, a conversation party, conversation contents, and a conversation time included in a conversation section determined by at least one of a start point and an end point selected by a user are identified. The conversation section including the identified conversation party, conversation contents, and conversation time are stored in the memory as a unit conversation.

Abstract: A method of providing a chat service between a first device and a plurality of second devices, includes: creating a first chat room configured to display visual signs transmitted and received between the first device and the plurality of second devices; receiving a command to select one or more of the plurality of second devices while the first chat room is displayed on the first device; selecting the one or more of the plurality of second devices based on the command; and creating a second chat room configured to display visual signs transmitted and received between the first device and the selected one or more of the plurality of second devices.