Abstract: Provided is a MEMS acoustic sensor including a substrate and a cavity, a back plate supported on the substrate and including a plurality of through-holes, at least one anchor projecting from the back plate toward the substrate, and a diaphragm supported by the at least one anchor and deformed by a sound wave introducing from the outside through the cavity, wherein no part of the deformed diaphragm comes into contact with the substrate.

Abstract: A method and a system for automatically generating multiple captions of an image are provided. A method for training an auto image caption generation model according to an embodiment of the present disclosure includes: generating a caption attention map by using an image; converting the generated caption attention map into a latent variable by projecting the caption attention map onto a latent space; deriving a guide map by using the latent variable; and training to generate captions of an image by using the guide map and the image. Accordingly, a plurality of captions describing various characteristics of an image and including various expressions can be automatically generated.

Abstract: An audio segmentation method based on an attention mechanism is provided. The audio segmentation method according to an embodiment obtains a mapping relationship between an “inputted text” and an “audio spectrum feature vector for generating an audio signal”, the audio spectrum feature vector being automatically synthesized by using the inputted text, and segments an inputted audio signal by using the mapping relationship. Accordingly, high quality can be guaranteed and the effort, time, and cost can be noticeably reduced through audio segmentation utilizing the attention mechanism.

Abstract: Provided is a display device including a substrate including a fingerprint area. The fingerprint area includes a first area and a second area. A light-emitting element layer is disposed on the substrate. The light-emitting element layer includes light-emitting elements. A sensor layer is disposed on the substrate, the sensor layer including photosensors. The light-emitting elements include a first light-emitting element arranged in the first area and a second light-emitting element arranged in the second area. During a fingerprint sensing period, the first light-emitting element emits light and the second light-emitting element does not emit light.

Abstract: Provided is a MEMS acoustic sensor including a substrate and a cavity, a back plate supported on the substrate and including a plurality of through-holes, at least one anchor projecting from the back plate toward the substrate, and a diaphragm supported by the at least one anchor and deformed by a sound wave introducing from the outside through the cavity, wherein no part of the deformed diaphragm comes into contact with the substrate.

Abstract: An audio synthesis method adapted to video characteristics is provided. The audio synthesis method according to an embodiment includes: extracting characteristics x from a video in a time-series way; extracting characteristics p of phonemes from a text; and generating an audio spectrum characteristic St used to generate an audio to be synthesized with a video at a time t, based on correlations between an audio spectrum characteristic St-1, which is used to generate an audio to be synthesized with a video at a time t?1, and the characteristics x. Accordingly, an audio can be synthesized according to video characteristics, and speech according to a video can be easily added.

Abstract: A foldable display device includes: a first substrate, a first thin-film transistor layer on the first substrate, a light emitting element layer on the first thin-film transistor layer, and a fingerprint sensor layer to receive light reflected by an external object. The fingerprint sensor layer includes: a second substrate under the first substrate, a second thin-film transistor layer under the second substrate, and a light receiving element layer under the second thin-film transistor layer.

Abstract: A display device is provided. The display device includes a first substrate, a first detection electrode on the first substrate, a first bank including an opening that exposes the first detection electrode, a photosensitive layer on the first detection electrode, a second detection electrode on the photosensitive layer, a first electrode on the second detection electrode, a second bank including an opening that exposes the first electrode, a light emitting layer on the first electrode, a second electrode on the light emitting layer, a first optical system between the second detection electrode and the first electrode, and a second optical system on the second electrode, wherein the first optical system and the second optical system overlap the photosensitive layer in a thickness direction of the display device.

Abstract: A sensor unit including driving electrodes and sensing electrodes, driving lines connected to the driving electrodes, sensing lines connected to the sensing electrodes, a driving signal output unit configured to sequentially apply driving signals to every P driving lines in a first driving mode, and a detector configured to receive detection signals from every Q sensing lines in the first driving mode, in which P and Q are positive integers.

Abstract: Disclosed is a method of providing a sign language video reflecting an appearance of a conversation partner. The method includes recognizing a speech language sentence from speech information, and recognizing an appearance image and a background image from video information. The method further comprises acquiring multiple pieces of word-joint information corresponding to the speech language sentence from joint information database, sequentially inputting the word-joint information to a deep learning neural network to generate sentence-joint information, generating a motion model on the basis of the sentence-joint information, and generating a sign language video in which the background image and the appearance image are synthesized with the motion model. The method provides a natural communication environment between a sign language user and a speech language user.

Abstract: A method and apparatus for recognizing a sign language or a gesture by using a three-dimensional (3D) Euclidean distance matrix (EDM) are disclosed. The method includes a two-dimensional (2D) EDM generation step for generating a 2D EDM including information about distances between feature points of a body recognized in image information by a 2D EDM generator, a 3D EDM generation step for receiving the 2D EDM and generating a 3D EDM by using a first deep learning neural network trained with training data in which input data is a 2D EDM and correct answer data is a 3D EDM by a 3D EDM generator, and a recognition step for recognizing a sign language or a gesture based on the 3D EDM.

Abstract: A display device includes: a display panel which includes a first area and a second area; and a sensor device which overlaps the first area in a thickness direction of the display panel. The display panel includes first sensor electrodes in the first area and second sensor electrodes in the second area, and a shape of a first sensor electrode of the first sensor electrodes is different from a shape of a second sensor electrode of the second sensor electrodes.

Abstract: Provided are a pen sensing unit and a display device. The pen sensing unit comprises a first pen line extending in a first direction; a pen scan line extending in a second direction intersecting the first direction; and a pen sensor member connected to the first pen line and the pen scan line and configured to sense input from a pen, wherein the pen sensor member comprises a pen sensor and a pen sensor circuit connected to a first end of the pen sensor, and the pen sensor circuit comprises a first thin film transistor.

Abstract: A touch panel includes a touch sensing circuit including a plurality of driving electrode lines and a plurality of sensing electrode lines, which intersect the driving electrode lines. The touch panel further includes a driving signal generation circuit that generates a plurality of driving signals and transmits the driving signals to the driving electrode lines, and a sensing circuit that receives a plurality of sensing signals from the sensing electrode lines. The driving signals have voltages that are not equal to zero during a first time zone, and the sum of driving signals generated at the same time is maintained to be equal to zero during the first time zone.

Abstract: The terminal includes a touch panel including first electrodes arranged in a first direction and second electrodes arranged in the second direction intersecting the first direction, a display panel attached to the touch panel and to display an image, and a processor to control the touch panel and the display panel. In a measuring mode, the processor is configured to: apply a driving signal to the touch panel, the touch panel being configured to form an electric field and/or magnetic field when the driving signal is applied, and determine body composition of a user based on a sensing signal output from the touch panel in accordance with eddy current, the eddy current being induced to a body of the user by the electric field and/or magnetic field formed around the touch panel.

Abstract: Deep learning-based automatic gesture recognition method and system are provided. The training method according to an embodiment includes: extracting a plurality of contours from an input image; generating training data by normalizing pieces of contour information forming each of the contours; and training an AI model for gesture recognition by using the generated training data. Accordingly, robust and high-performance automatic gesture recognition can be performed, without being influenced by an environment and a condition even while using less training data.

Abstract: A washing machine is disclosed. The washing machine include a first housing, a second housing arranged above the first housing, a first tub covered by the first housing, a second tub covered by the second housing, and a first detergent supplier arranged to supply a detergent to the first tub, a second detergent supplier having a first supply assembly arranged to supply a detergent to the second tub by siphonage when water for laundry is being supplied, wherein the first supply assembly comprises a first valve device arranged in a detergent supply path, in which a detergent is moved to the second tub, for preventing water for laundry from flowing backwards by buoyancy.

Abstract: There are provided a fingerprint sensor and a display device including the same. The fingerprint sensor includes a base substrate in which a first sensing region and a second sensing region are defined on a plane, first photo sensors disposed on the first sensing region of the base substrate, and second photo sensors disposed on the second sensing region of the base substrate. The first sensing region has an area larger than that of the second sensing region, and the first photo sensor has a width wider than that of the second photo sensor.

Abstract: A method and a system for automatic image caption generation are provided. The automatic image caption generation method according to an embodiment of the present disclosure includes: extracting a distinctive attribute from example captions of a learning image; training a first neural network for predicting a distinctive attribute from an image, by using a pair of the extracted distinctive attribute and the learning image; inferring a distinctive attribute by inputting the learning image to the trained first neural network; and training a second neural network for generating a caption of an image by using a pair of the inferred distinctive attribute and the learning image. Accordingly, a caption well indicating a feature of a given image is automatically generated, such that an image can be more exactly explained and a difference from other images can be clearly distinguished.

Abstract: A thin film transistor array panel includes a substrate, a light blocking film disposed on the substrate, a buffer layer covering the light blocking film, and a channel region disposed on the buffer layer. A source region and a drain region are disposed in the same layer as the channel region. A gate insulating layer is disposed on the channel region, and a gate electrode overlaps the channel region, with the gate insulating layer interposed between the gate electrode and the channel region. A passivation layer is disposed on the gate electrode, the source region, the drain region, and the buffer layer. A source electrode and a drain electrode are disposed on the passivation layer, wherein the channel region, the source region, and the drain region comprise an oxide semiconductor, and wherein a carrier concentration of the source region and the drain region is larger than in the channel region.