Abstract: The present invention relates to a convolutional artificial neural network-based image and video recognition system, and a method therefor. The recognition system comprises: a mobile device for performing lower layer analysis, transmitting a user question to a server, receiving an answer thereto from the server, and managing the same; the server connected to the mobile device over a network so as to perform data processing in respective neural network layers corresponding to middle and upper layers of a convolutional artificial neural network, and register, analyze, search for, and classify a particular object (image) and video; and a retriever for comparing an FC layer value for artificial neural network processing of an image and video transmitted by the mobile device with an FC layer value for artificial neural network processing of an image and video registered by a content provider.

Abstract: An integrated circuit device includes a sense amplifier configured to sense a voltage change of a bit line, wherein the sense amplifier includes: a sense amplifier unit connected to the bit line and a complementary bit line, configured to sense the voltage change of the bit line in response to a control signal, configured to adjust voltages of a sensing bit line and a complementary sensing bit line based on the sensed voltage change, and including a first PMOS transistor and a first NMOS transistor; and a first offset canceling unit connecting the bit line to the complementary sensing bit line in response to an offset canceling signal, and including a first offset canceling transistor arranged between the first NMOS transistor and the first PMOS transistor, wherein the first offset canceling transistor shares a common impurity region with the first NMOS transistor.

Abstract: A 3D nanopore device for characterizing biopolymer molecules includes a first selecting layer having a first axis of selection. The device also includes a second selecting layer disposed adjacent the first selecting layer and having a second axis of selection orthogonal to the first axis of selection. The device further includes an third electrode layer disposed adjacent the second selecting layer, such that the first selecting layer, the second selecting layer, and the third electrode layer form a stack of layers along a Z axis and define a plurality of nanopore pillars.

Abstract: A method of detecting a target includes generating an image pyramid based on an image on which a detection is to be performed; classifying candidate areas in the image pyramid using a cascade neural network; and determining a target area corresponding to a target included in the image based on the plurality of candidate areas, wherein the cascade neural network includes a plurality of neural networks, and at least one neural network among the neural networks includes parallel sub-neural networks.

Abstract: A 3D nanopore device for characterizing biopolymer molecules includes a first selecting layer having a first axis of selection. The device also includes a second selecting layer disposed adjacent the first selecting layer and having a second axis of selection orthogonal to the first axis of selection. The device further includes an third electrode layer disposed adjacent the second selecting layer, such that the first selecting layer, the second selecting layer, and the third electrode layer form a stack of layers along a Z axis and define a plurality of nanopore pillars.

Abstract: The present disclosure provides a drug discharge assembly including a drive piece linearly reciprocated along one direction, a rotating unit being in contact with an end of the drive piece and rotating in one direction according to a linear reciprocating motion of the drive piece, a flexible tube disposed adjacent to the rotating unit and having a curved section at least partially extending in a circumferential direction, and a pressing unit mounted on the rotating unit and rotating while pressing the tube according to a rotation of the rotating unit, and a drug injection device comprising the same.

Abstract: A lightened neural network method and apparatus. The neural network apparatus includes a processor configured to generate a neural network with a plurality of layers including plural nodes by applying lightened weighted connections between neighboring nodes in neighboring layers of the neural network to interpret input data applied to the neural network, wherein lightened weighted connections of at least one of the plurality of layers includes weighted connections that have values equal to zero for respective non-zero values whose absolute values are less than an absolute value of a non-zero value. The lightened weighted connections also include weighted connections that have values whose absolute values are no greater than an absolute value of another non-zero value, the lightened weighted connections being lightened weighted connections of trained final weighted connections of a trained neural network whose absolute maximum values are greater than the absolute value of the other non-zero value.

Abstract: An integrated circuit device includes a sense amplifier configured to sense a voltage change of a bit line, wherein the sense amplifier includes: a sense amplifier unit connected to the bit line and a complementary bit line, configured to sense the voltage change of the bit line in response to a control signal, configured to adjust voltages of a sensing bit line and a complementary sensing bit line based on the sensed voltage change, and including a first PMOS transistor and a first NMOS transistor; and a first offset canceling unit connecting the bit line to the complementary sensing bit line in response to an offset canceling signal, and including a first offset canceling transistor arranged between the first NMOS transistor and the first PMOS transistor, wherein the first offset canceling transistor shares a common impurity region with the first NMOS transistor.

Abstract: A dielectrically-enhanced polyethylene formulation made from (A) a polyethylene-nucleation agent, a dielectrically-enhanced (B) a multimodal ultra-high density polyethylene (multimodal uHDPE) composition, and, optionally, (C) a low density polyethylene (LDPE) polymer. A method of making same, an insulated electrical device comprising same, and a method of using same.

Abstract: A method of detecting a pathogen uses a 3D nanochannel device having top and bottom chambers, and a plurality of nanochannels. The method also includes functionalizing a nanochannel by coupling an oligonucleotide probe to an inner surface thereof. The method further includes adding a lysis buffer and patient sample to the top chamber. Moreover, the method includes extracting an oligonucleotide from the patient sample. In addition, the method includes placing top and bottom electrodes in the top and bottom chambers respectively and applying an electrophoretic bias therethrough. The method also includes applying a selection bias across first and second gating nanoelectrodes to direct flow of the oligonucleotide through the nanochannel. Moreover, the method includes applying a sensing bias through a sensing nanoelectrode. In addition, the method includes detecting an output current from the sensing nanoelectrode, and analyzing the output current from the sensing nanoelectrode to detect the oligonucleotide.

Abstract: Provided are a hybrid insole with a multi-shock absorbing pad and a method for fabricating the hybrid insole. The hybrid insole includes an insole body including at least one sheet layer that is formed of an elastic material and has a shape of a human sole; a first absorbing pad means including first to fourth absorbing pad portions attached to the insole body corresponding to a metatarsal region, a tarsal bone region, and a calcaneus region of the human foot, and first to third connecting portions selectively connecting the first to fourth absorbing pad portions; and a second absorbing pad means including a fifth absorbing pad portion attached to the insole body corresponding to the metatarsal region and a phalange region of the human foot. The hybrid insole is configured such that absorbing pads of different structures are disposed to correspond to the bone structure for each region of the foot and the shape of an arch, thus distributing and absorbing the weight load of the human body.

Abstract: A method of manufacturing and using a nanofluidic NAND transistor sensor array scheme including a plurality of nanopore channel pillars, a plurality of respective fluidic channels, a plurality of gate electrodes, a top chamber, and a bottom chamber includes placing a sensor substrate in an electrolyte solution comprising biomolecules and DNA. The method also includes placing first and second electrodes in the electrolyte solution (Vpp and Vss of the nanofluidic NAND transistor); forming the nanopore channel pillars; placing the gate electrodes and gate insulators in respective walls of the nanopore channel pillars; applying an electrophoretic bias in the first and second electrodes; applying a bias in the gate electrodes; detecting a change in an electrode current in the electrolyte solution caused by a change in a gate voltage; and detecting a change in a surface charge in nanopore channel electrodes in the respective fluidic channels.

Abstract: A lightened neural network method and apparatus. The neural network apparatus includes a processor configured to generate a neural network with a plurality of layers including plural nodes by applying lightened weighted connections between neighboring nodes in neighboring layers of the neural network to interpret input data applied to the neural network, wherein lightened weighted connections of at least one of the plurality of layers includes weighted connections that have values equal to zero for respective non-zero values whose absolute values are less than an absolute value of a non-zero value. The lightened weighted connections also include weighted connections that have values whose absolute values are no greater than an absolute value of another non-zero value, the lightened weighted connections being lightened weighted connections of trained final weighted connections of a trained neural network whose absolute maximum values are greater than the absolute value of the other non-zero value.

Abstract: A 3D nanopore device for characterizing biopolymer molecules includes a first selecting layer having a first axis of selection. The device also includes a second selecting layer disposed adjacent the first selecting layer and having a second axis of selection orthogonal to the first axis of selection. The device further includes an third electrode layer disposed adjacent the second selecting layer, such that the first selecting layer, the second selecting layer, and the third electrode layer form a stack of layers along a Z axis and define a plurality of nanopore pillars.

Abstract: A gaze estimation method and apparatus is disclosed. The gaze estimation method includes obtaining an image including an eye region of a user, extracting, from the obtained image, a first feature of data, obtaining a second feature of data used for calibration of a neural network model, and estimating a gaze of the user using the first feature and the second feature.

Abstract: An integrated circuit packaging system and method of manufacture thereof includes: a substrate with internal circuitry between a substrate top side, a substrate bottom side, and vertical sides; an integrated circuit coupled to the internal circuitry; a molded package body formed directly on the integrated circuit and the substrate top side of the substrate; and a conductive conformal shield structure applied directly on the molded package body, the vertical sides, and to extend below the substrate bottom side coupled to the internal circuitry.

Abstract: Disclosed is a liveness test method and apparatus. A liveness test apparatus determines a pre-liveness score based on a plurality of sub-images acquired from an input image, determines a post-liveness score based on a recognition model for recognizing an object included in the input image, and determines a liveness of the object based on any one or any combination of the pre-liveness score and the post-liveness score.

Abstract: Provided is a liveness verification method and device. A liveness verification device acquires a first image and a second image, and select one or more liveness models based on respective analyses of the first image and the second image, including analyses based on an object part being detected in the first image and/or the second image, and to verify, using the selected one or more liveness models, a liveness of the object based on the first image and/or the second image. The first image may be a color image and the second image may be an Infrared image.

Abstract: A gaze estimation method and apparatus is disclosed. The gaze estimation method includes obtaining an image including an eye region of a user, extracting, from the obtained image, a first feature of data, obtaining a second feature of data used for calibration of a neural network model, and estimating a gaze of the user using the first feature and the second feature.

Abstract: A facial verification method and apparatus is disclosed. The facial verification method includes detecting a face region in an input image, determining whether the detected face region represents a partial face, in response to a determination that the detected face region represents the partial face, generating a synthesized image by combining image information of the detected face region and reference image information, performing a verification operation with respect to the synthesized image and predetermined first registration information, and indicating whether facial verification of the input image is successful based on a result of the performed verification operation.