Abstract: A data augmentation method includes inputting multiple images constituting a mini-batch into an encoder and extracting features for respective images of the multiple images, inputting the extracted features of the respective images into a pre-trained first aggregator and second aggregator and separating the extracted features into object features, each being a feature of an object portion of each image, and background features, each being a feature of a background portion of each image, inputting the object feature and background feature of each of the images into a shuffler and shuffling either the object features or the background features within the mini-batch, generating a synthetic feature by synthesizing the shuffled feature and a non-shuffled feature among the object feature and the background feature in a synthesis unit, and generating a data-augmented image based on the synthetic feature.

Abstract: A data diversity augmentation method includes inputting a sentence into an encoder and extracting a feature vector for the sentence, inputting the extracted feature vector into an attribute classifier to form a decision boundary for the feature vector, and moving the extracted feature vector based on the decision boundary to generate a transformed feature vector and inputting the transformed feature vector into a decoder to restore a transformed sentence for the sentence.

Abstract: An apparatus for constructing captioning data according to an embodiment is provided with one or more processors and a memory storing one or more programs executed by the one or more processors, and includes an input module configured to acquire input sentences for an image for which captions are to be acquired, and a generation module configured to generate a captioning dataset by paraphrasing and translating the input sentences to generate a plurality of paraphrased and translated sentences and using the plurality of generated paraphrased and translated sentences as a set.

Abstract: A method for detecting profanity according to an embodiment is performed in a computing device that includes one or more processors and a memory storing one or more programs executed by the one or more processors includes acquiring sequence data converted from a sentence input by a user, and generating embedded data containing one or more tokens by embedding the acquired sequence data, and training a neural network model to output information on whether the sentence contains profanity by inputting the generated embedded data into the neural network model.

Abstract: A display device and an electronic device using the same are disclosed. A display device includes a display panel configured to display images through pixels in a display area, a touch driver circuit configured to detect a touch of a user through a touch sensor of the display panel and generate touch coordinate data, a plurality of data driver circuits configured to apply data voltage to the pixels in the display area, a display driver circuit configured to control driving timing of the plurality of data driver circuits and image display timing of the pixels, and a noise blocking film covering a driver including the plurality of data driver circuits, the display driver circuit, and the touch driver circuit, and configured to block electromagnetic waves from the driver.

Abstract: A document classification method is performed in a computing device equipped with one or more processors and a memory storing one or more programs executed by the one or more processors. The document classification method includes acquiring sequence data obtained by digitizing a document, and embedding the acquired sequence data to produce embedded data including one or more tokens, generating fuzzy-pruned data pruned by performing token pruning the embedded data, which is a set of tokens, based on fuzzy theory, producing combined tokens by combining tokens of the pruned fuzzy-pruned data with seed tokens of seed data, and generating combined data as a set of the combined tokens, and producing one classification value based on the combined data.

Abstract: Provided is a method and apparatus for registering metadata regarding a drone image. The apparatus for registering metadata regarding a drone image acquires captured data, and parses metadata from the captured data. The apparatus for registering metadata regarding a drone image generates new metadata using additional information and the parsed metadata, and registers the generated new metadata and the captured data to generate new captured data.

Abstract: A display device includes: a display panel; a panel cover disposed on a back surface of the display panel and including one or more alignment marks located on a back surface of the panel cover; a printed circuit board disposed on the back surface of the panel cover and including a ground area located around the one or more the alignment marks; and a protective member disposed on the printed circuit board and including a transparent portion at least partially overlapping the ground area in a plan view.

Abstract: Proposed are an apparatus and a method for simulating charge and discharge energy in an intermittent computing system. Specifically, an operation method for simulating power instability in intermittent computing includes receiving at least one selected from a group of energy storage and SBC standard information of an intermittent computing system, energy consumption information for an application program of the intermittent computing system, and illuminance information, and generating an intermittent computing charge and discharge energy scenario model on the basis of the received information, and performing intermittent computing energy simulation on the basis of the generated model, and identifying whether a result of the simulation is appropriate.

Abstract: A metamaterial structure according to one embodiment of the present disclosure is formed in a metal pattern that allows microwave transmittance to approach the transmittance of air in a specific frequency band of microwaves. In this case, the metal pattern is provided in an electrically interconnected form to perform a heating function.

Abstract: A metamaterial structure according to one embodiment of the present disclosure is formed in a metal pattern that allows microwave transmittance to approach the transmittance of air in a specific frequency band of microwaves. In this case, the metal pattern is provided in an electrically interconnected form to perform a heating function.

Abstract: A metamaterial structure according to one embodiment of the present disclosure is formed in a metal pattern that allows microwave transmittance to approach the transmittance of air in a specific frequency band of microwaves. In this case, the metal pattern is provided in an electrically interconnected form to perform a heating function.

Abstract: A metamaterial structure according to one embodiment of the present disclosure is formed in a metal pattern that allows microwave transmittance to approach the transmittance of air in a specific frequency band of microwaves. In this case, the metal pattern is provided in an electrically interconnected form to perform a heating function.

Abstract: Provided is a method and apparatus for registering metadata regarding a drone image. The apparatus for registering metadata regarding a drone image acquires captured data, and parses metadata from the captured data. The apparatus for registering metadata regarding a drone image generates new metadata using additional information and the parsed metadata, and registers the generated new metadata and the captured data to generate new captured data.

Abstract: A resource resettable deep neural network accelerator according to an embodiment of the present disclosure includes: a memory layer including a scratchpad memory layer configured to divide deep neural network parameter data (hereinafter, data) in an external memory layer into a plurality of tiles and to load the divided tiles, and a register file memory layer configured to load tiled data of the scratchpad memory layer; and a plurality of cores configured to process an inference operation for the data loaded in the register file memory layer, wherein the memory layer includes a virtual tiling layer added to a certain location for loading the tiled data from a previous memory layer so as to correspond to a specific tiling size.

Abstract: An embodiment of the present invention provides a manufacturing method of an herbal medicinal tablet formulation for treating obesity which is prescribed based on Sasang constitutional medicine, comprises manufacturing a concentrated ephedra powder agent so that an ephedrine content is 3.0-4.0%; determining a weight of the concentrated ephedra powder agent according to each constitution of Sasang constitutional medicine, and manufacturing a side effect-preventing powder agent for each constitution to prevent and suppress side effects according to constitution of Sasang constitutional medicine with respect to the weight of the concentrated ephedra powder agent; mixing the concentrated ephedra powder agent and the side effect-preventing powder agent with a variance of a weight ratio therebetween in consideration of weight, obesity, constitution of Sasang constitutional medicine, and side effects; and tableting the mixture of the prepared ephedra powder agent and side effect-preventing powder agent.

Abstract: An embodiment of the present invention provides an herbal medicinal tablet formulation for treating obesity which can be prescribed based on Sasang constitutional medicine, including: an ephedra powder agent of which ephedrine content is concentrated to be 3.0 to 4.0%, which is greater than 0.5 to 2.5% when it is in a natural state, and of which the ephedrine content is maintained constant at 3.0 to 4.0%; and a side-effect-preventing powder agent for each constitution prescribed based on Sasang constitutional medicine in order to prevent a side effect of the ephedra powder agent, wherein a weight ratio of the ephedra powder agent and the side-effect-preventing powder agent is 8-10:1 to 9-11:1.

Abstract: Provided is a light source.

Abstract: Disclosed is a substrate treating apparatus. The substrate treating apparatus includes a chamber having a treatment space in the interior thereof, a support unit configured to support a substrate in the treatment space, a gas supply unit configured to supply a gas into the treatment space, and a plasma generating unit configured to generate plasma from the gas, wherein the support unit includes an electrostatic chuck including an upper body having a support surface that suctions the substrate and a lower body extending from the upper body to a lower side, wherein the lower body has an extension part extending laterally from the upper body, a focus ring disposed on the extension part of the electrostatic chuck, and a metallic ring provided between the upper body of the electrostatic chuck and the focus ring and configured to control plasma in an extreme edge of the substrate.

Abstract: An embodiment of the present invention provides a manufacturing method of an herbal medicinal tablet formulation for treating obesity which can be prescribed based on Sasang constitutional medicine, comprises manufacturing a concentrated and quantified ephedra powder agent so that an ephedrine content of ephedra, which differs from 0.5-2.5% depending on a producing district and collecting period, is 3.0-4.0%, determining a weight according to each constitution of Sasang constitutional medicine of the concentrated and quantified ephedra powder agent of which the ephedrine content is 3.0-4.