Abstract: The present invention relates to a multi-modality system for recommending multiple items using an interaction and a method of operating the same. The multi-modality system includes an interaction data preprocessing module that preprocesses an interaction data set and converts the preprocessed interaction data set into interaction training data; an item data preprocessing module that preprocesses item information data and converts the preprocessed item information data into item training data; and a learning module that includes a neural network model that is trained using the interaction training data and the item training data and outputs a result including a set of recommended items using a conversation context with a user as input.

Abstract: A method including clustering second point clouds comprised in second point cloud data of a second timepoint, obtaining output data corresponding to one or more second clusters generated as a result of the clustering by applying the second point clouds to a pretrained neural network, determining whether at least one of the one or more second clusters is a target object based on a reliability factor, tracking first clusters corresponding to the target object in first point cloud data of one or more first timepoints prior to the second timepoint, and training of the pretrained neural network based on the output data corresponding to the first clusters and the first point clouds for each first cluster determined according to a result of the tracking.

Abstract: Disclosed is a sign apparatus for displaying a sign for traffic or a vehicle, where the sign apparatus includes a display configured to display the sign and to permit a radar signal radiated from a source to pass, a filter configured to attenuate a signal of at least one frequency in the radar signal passing through the display and to allow a signal of a remaining frequency to pass, and a reflector configured to reflect the signal passing through the filter, wherein the reflected signal includes encoded identification information of the sign.

Abstract: A control apparatus includes a Wi-Fi communication module, a processor, a controller that controls a component of a vehicle, and a memory, wherein the memory stores at least one piece of position information corresponding to a position in the vehicle, and at least one registered pattern of a gesture, face, or shape, the Wi-Fi communication module transmits a transmission signal to the position corresponding to the at least one piece of position information by a Wi-Fi communication method at a preset frequency and receives a reflection signal reflected from an object in response to the transmission signal, the processor acquires an input pattern of a gesture, face, or shape of the object by analyzing the reflection signal, and retrieves a registered pattern matching the input pattern from the at least one registered pattern, and the controller controls the component according to an instruction corresponding to the retrieved registered pattern.

Abstract: A method of predicting a movement of a cut-in object, including separating a Doppler velocity corresponding to at least one wheel of the cut-in object from radar data based on a velocity of the cut-in object, determining a position of the at least one wheel of the cut-in object based on the Doppler velocity, generating at least one of first movement information related to a horizontal movement of the cut-in object or second movement information related to a moving direction of the cut-in object based on the position of the at least one wheel, and determining a position of the cut-in object based on at least one of the first movement information or the second movement information.

Abstract: A method and device for object recognition for information collected from a sensor of a moving object are disclosed. The method may include identifying a target with a confidence level of recognition accuracy less than a threshold, based on first sensing information collected from a moving object, obtaining second sensing information on the target from an interworking moving object, synchronizing the first sensing information and the second sensing information based on time information included in the first sensing information and time information included in the second sensing information, and training a neural network for estimating recognition information of the target from the first synchronized sensing information by using information on the target included in the second synchronized sensing information as ground truth data.

Abstract: A convertible hat is disclosed. The convertible hat of the present invention comprises: a band (110) coupled to the head of a user; and a brim (120) provided horizontally in a certain area on the front surface of the band (110) to block sunlight. Furthermore, the convertible hat of the present invention preferably includes a pair of shape transformation crowns (130, 130a) that each have one end coupled to the band (110) and the other end spread out over the band (110), and are positioned in either a cap position in which the crowns cover the head of the user, or a sun cap position in which the crowns extend from the side surface of the brim (120) to increase the sun blocking area of the brim (120). Accordingly, a single convertible hat can be used by transforming into the two forms of a cap and a sun cap.

Abstract: A display device includes a data conductive layer including a first power line, a passivation layer with a first opening exposing the first power line, a via layer with a second opening partially overlapping the first opening, a pixel electrode on the via layer, a connection electrode in the first and second openings, a pixel-defining film with an opening overlapping the second opening, a light-emitting layer on the pixel-defining film, the pixel electrode and the connection electrode, and a common electrode connected to the first power line. The data conductive layer includes a data base layer, a data main metal layer, and a data capping layer, the first power line includes a wire connection structure, in which the data main metal layer is recessed from sides of the data capping layer, and the common electrode is connected to the data main metal layer in the wire connection structure.

Abstract: A transistor array substrate includes a substrate, an active layer disposed on the substrate and including a channel region, a source region and a drain region, a gate insulating layer disposed on a part of the active layer, a gate electrode overlapping the channel region of the active layer and included in an electrode conductive layer which is disposed on the gate insulating layer, a source electrode included in the electrode conductive layer and in contact with a part of the source region of the active layer, and a drain electrode included in the electrode conductive layer and in contact with a part of the drain region of the active layer. The active layer includes an oxide semiconductor including crystals and is disposed as an island shape excluding a hole in a plan view.

Abstract: A display device includes a substrate, a first conductive layer on the substrate, the first conductive layer including a data signal line, a first insulating layer on the first conductive layer, a semiconductor layer on the first insulating layer, the semiconductor layer including a first semiconductor pattern, a second insulating layer on the semiconductor layer, and a second conductive layer on the second insulating layer, the second conductive layer including a gate electrode disposed to overlap the first semiconductor pattern, a transistor first electrode disposed to overlap a part of the first semiconductor pattern, wherein the transistor first electrode is electrically connected to the data signal line through a contact hole that penetrates the first and second insulating layers, and a transistor second electrode disposed to overlap another part of the first semiconductor pattern.

Abstract: A radar signal processing method includes: extracting a first chirp sequence signal of a first carrier frequency and a second chirp sequence signal of a second carrier frequency from a radar signal received through an array antenna in a radar sensor; generating a first range-Doppler map by performing frequency conversion on the first chirp sequence signal; detecting a first target cell corresponding to a first target in the first range-Doppler map; determining a first ambiguous Doppler velocity of the first target based on first frequency information of the first target cell; determining a first range of an unambiguously measurable Doppler velocity through the first chirp sequence signal; estimating second ambiguous Doppler velocities based on the first ambiguous Doppler velocity and the first range; and determining a Doppler velocity of the first target by performing partial frequency conversion on the second chirp sequence signal based on the second ambiguous Doppler velocities.

Abstract: Disclosed is a liveness test method and liveness test apparatus. The liveness test method includes determining a presence of a subject using a radar sensor, performing a first liveness test on the subject based on radar data obtained by the radar sensor, in response to the subject being present, acquiring image data of the subject using an image sensor, in response to a result of the first liveness test satisfying a first condition, and performing a second liveness test on the subject based on the image data.

Abstract: A radar signal processing method includes: extracting a first chirp sequence signal of a first carrier frequency and a second chirp sequence signal of a second carrier frequency from a radar signal received through an array antenna in a radar sensor; generating a first range-Doppler map by performing frequency conversion on the first chirp sequence signal; detecting a first target cell corresponding to a first target in the first range-Doppler map; determining a first ambiguous Doppler velocity of the first target based on first frequency information of the first target cell; determining a first range of an unambiguously measurable Doppler velocity through the first chirp sequence signal; estimating second ambiguous Doppler velocities based on the first ambiguous Doppler velocity and the first range; and determining a Doppler velocity of the first target by performing partial frequency conversion on the second chirp sequence signal based on the second ambiguous Doppler velocities.

Abstract: A convertible hat is disclosed. The convertible hat of the present invention comprises: a band (110) coupled to the head of a user; and a brim (120) provided horizontally in a certain area on the front surface of the band (110) to block sunlight. Furthermore, the convertible hat of the present invention preferably includes a pair of shape transformation crowns (130, 130a) that each have one end coupled to the band (110) and the other end spread out over the band (110), and are positioned in either a cap position in which the crowns cover the head of the user, or a sun cap position in which the crowns extend from the side surface of the brim (120) to increase the sun blocking area of the brim (120). Accordingly, a single convertible hat can be used by transforming into the two forms of a cap and a sun cap.

Abstract: Disclosed is a liveness test method and liveness test apparatus. The liveness test method includes determining a presence of a subject using a radar sensor, performing a first liveness test on the subject based on radar data obtained by the radar sensor, in response to the subject being present, acquiring image data of the subject using an image sensor, in response to a result of the first liveness test satisfying a first condition, and performing a second liveness test on the subject based on the image data.

Abstract: A control apparatus includes a Wi-Fi communication module, a processor, a controller that controls a component of a vehicle, and a memory, wherein the memory stores at least one piece of position information corresponding to a position in the vehicle, and at least one registered pattern of a gesture, face, or shape, the Wi-Fi communication module transmits a transmission signal to the position corresponding to the at least one piece of position information by a Wi-Fi communication method at a preset frequency and receives a reflection signal reflected from an object in response to the transmission signal, the processor acquires an input pattern of a gesture, face, or shape of the object by analyzing the reflection signal, and retrieves a registered pattern matching the input pattern from the at least one registered pattern, and the controller controls the component according to an instruction corresponding to the retrieved registered pattern.

Abstract: An operating method of a radar sensor includes: performing object classification on a detected target; comparing a confidence score of a result of the object classification with a threshold value; determining whether high-resolution information on the detected target is to be generated based on a result of the comparison; in response to a determination that the high-resolution information is to be generated, selecting one or more of a plurality of analog-to-digital converters (ADCs) of the radar sensor, and determining control information for changing a setting of the selected one or more ADCs; changing the setting of the selected one or more ADCs based on the determined control information; generating first intermediate frequency signals based on respective first reception signals and a frequency modulated signal, the first reception signals being signals received through reception antennas after a first transmission signal output by the radar sensor is reflected by the detected target, and the first transmi

Abstract: A method with radio detection and ranging (radar) data processing may include: obtaining, by a radar sensor, input radar data; and generating, using a resolution increase model, output radar data from the input radar data and reference data, wherein the output radar data has a resolution greater than a resolution of the input radar data.

Abstract: A control apparatus includes a Wi-Fi communication module, a processor, a controller that controls a component of a vehicle, and a memory, wherein the memory stores at least one piece of position information corresponding to a position in the vehicle, and at least one registered pattern of a gesture, face, or shape, the Wi-Fi communication module transmits a transmission signal to the position corresponding to the at least one piece of position information by a Wi-Fi communication method at a preset frequency and receives a reflection signal reflected from an object in response to the transmission signal, the processor acquires an input pattern of a gesture, face, or shape of the object by analyzing the reflection signal, and retrieves a registered pattern matching the input pattern from the at least one registered pattern, and the controller controls the component according to an instruction corresponding to the retrieved registered pattern.

Abstract: A focused ultrasonic transducer includes a piezoelectric substrate having a first face and a second face, a back metal layer disposed over the first face, and a patterned metal layer disposed over the second face. The patterned metal layer includes a first plurality of concentric ring electrodes wherein each of the first plurality of concentric ring electrodes are wired to be individually accessible. A controller actuates a subset of the concentric ring electrodes such that electrical control of focal length is achieved by selecting a group of electrodes to actuate so that acoustic waves generated from selected electrodes arrive at a desired focal length in-phase and interfere constructively to create a focal spot of high acoustic intensity. The patterned metal layer optionally includes a first central electrode that is surrounded by the first plurality of concentric ring electrodes.