Abstract: A video encoding/decoding apparatus according to the present invention acquires motion vector refinement information, performs motion compensation on the basis of a motion vector of a current block, refines the motion vector of the current block using at least one or both of the motion vector refinement information and the output of the motion compensation, and performs motion compensation using the refined motion vector.

Abstract: Provided is a device for upscaling resolution based on a slice image, in which a low-resolution image is divided into a plurality of slice images so that a high-resolution image is generated. The device includes a convolution operation unit configured to convert a low-resolution input slice image into a high-resolution output slice image using a convolutional neural network. The convolutional neural network includes a cascading block configured to perform a convolution operation using a convolution filter having a predetermined size and a residual operation on an input feature map generated from the low-resolution input slice image to generate an output feature map, and an upscaling block configured to upscale the output feature map to generate the high-resolution output slice image.

Abstract: A control method of a fastening tool for fastening a component part includes photographing a fastening portion of the component part through a camera portion mounted on the fastening tool. The control method includes pre-processing that rotates an input image, which has been photographed by the camera portion, to match with representative model image. The control method includes estimating the fastening portion through an inference process through a convolutional neural network (CNN) algorithm-based image classification work on a video input image of a same fastening portion finished with the pre-processing work. The control method also includes setting a torque value that is matched with a recognized fastening portion when a probability value of the fastening portion in the inference process exceeds a predetermined reference ratio.

Abstract: A fastening tool system for fastening a component part by using a supplied power includes: a camera portion mounted on a fastening tool to photograph a fastening portion of the component part; an image input portion for inputting an image photographed from the camera portion when a distance with the fastening portion measured through a distance measurement sensor is within a predetermined distance; a communication portion for transmitting data by wireless communication; and a fastening portion determining portion that recognizes the fastening portion through machine learning algorithm-based image classification work for an input image of the camera portion and sets a torque value matched with the recognized fastening portion.

Abstract: A video encoding/decoding apparatus according to the present invention acquires motion vector refinement information, performs motion compensation on the basis of a motion vector of a current block, refines the motion vector of the current block using at least one or both of the motion vector refinement information and the output of the motion compensation, and performs motion compensation using the refined motion vector.

Abstract: A video encoding/decoding apparatus according to the present invention acquires motion vector refinement information, performs motion compensation on the basis of a motion vector of a current block, refines the motion vector of the current block using at least one or both of the motion vector refinement information and the output of the motion compensation, and performs motion compensation using the refined motion vector.

Abstract: A method for controlling a mobile communication device using a user's distance specifying algorithm, the method comprises a first process of receiving an input image from a camera to generate first data; a second process of determining whether there is a candidate group that can be determined as a face using the first data and second data including a user's distance information; a third process of extracting, if there is a candidate group that can be determined as a face, from a plurality of third data including location information of the candidate group, fourth data having a minimum distance between a center of the first data and a center of the third data; and a fourth process of comparing the fourth data with a predetermined user's distance to determine whether it has a value within a tolerance range.

Abstract: Systems and methods for detecting eye information are described. According to one embodiment, a detecting device includes a determination unit, a generation unit, and a detection unit. The determination unit is configured to determine whether a user wears glasses from an input image by using a first neural network including a plurality of filters. The generation unit is configured to generate a face image in which the glasses are removed from the input image by using a learned generator when a glasses-wearing signal is received from the determination unit. The detection unit is configured to generate an eye image by receiving the input image from the determination unit or receiving the face image from the generation unit and detect a pupil from the eye image using a second neural network including a plurality of filters.

Abstract: A pupil detection device includes: a filter weight learner configured to generate a target image on the basis of pupil coordinate data acquired from a learning image and learn weights of a plurality of filters in order to generate a filtered image, by filtering the learning image using the plurality of filters, that is within a predetermined reference range of the target image; a split image generator configured to generate a pupil region split image for an input image using the plurality of filters having the learned weights; and a pupil coordinate detector configured to remove noise of the pupil region split image, select at least one of a plurality of pixels from which noise is removed, and detect pupil coordinates.

Abstract: A video encoding/decoding apparatus according to the present invention acquires motion vector refinement information, performs motion compensation on the basis of a motion vector of a current block, refines the motion vector of the current block using at least one or both of the motion vector refinement information and the output of the motion compensation, and performs motion compensation using the refined motion vector.

Abstract: A video encoding/decoding apparatus according to the present invention acquires motion vector refinement information, performs motion compensation on the basis of a motion vector of a current block, refines the motion vector of the current block using at least one or both of the motion vector refinement information and the output of the motion compensation, and performs motion compensation using the refined motion vector.

Abstract: A video encoding/decoding apparatus according to the present invention acquires motion vector refinement information, performs motion compensation on the basis of a motion vector of a current block, refines the motion vector of the current block using at least one or both of the motion vector refinement information and the output of the motion compensation, and performs motion compensation using the refined motion vector.

Abstract: A method for controlling a mobile communication device using a user's distance specifying algorithm, the method comprises a first process of receiving an input image from a camera to generate first data; a second process of determining whether there is a candidate group that can be determined as a face using the first data and second data including a user's distance information; a third process of extracting, if there is a candidate group that can be determined as a face, from a plurality of third data including location information of the candidate group, fourth data having a minimum distance between a center of the first data and a center of the third data; and a fourth process of comparing the fourth data with a predetermined user's distance to determine whether it has a value within a tolerance range.

Abstract: Provided is a device for upscaling resolution based on a slice image, in which a low-resolution image is divided into a plurality of slice images so that a high-resolution image is generated. The device includes a convolution operation unit configured to convert a low-resolution input slice image into a high-resolution output slice image using a convolutional neural network. The convolutional neural network includes a cascading block configured to perform a convolution operation using a convolution filter having a predetermined size and a residual operation on an input feature map generated from the low-resolution input slice image to generate an output feature map, and an upscaling block configured to upscale the output feature map to generate the high-resolution output slice image.

Abstract: Systems and methods for detecting eye information are described. According to one embodiment, a detecting device includes a determination unit, a generation unit, and a detection unit. The determination unit is configured to determine whether a user wears glasses from an input image by using a first neural network including a plurality of filters. The generation unit is configured to generate a face image in which the glasses are removed from the input image by using a learned generator when a glasses-wearing signal is received from the determination unit. The detection unit is configured to generate an eye image by receiving the input image from the determination unit or receiving the face image from the generation unit and detect a pupil from the eye image using a second neural network including a plurality of filters.

Abstract: A pupil detection device includes: a filter weight learner configured to generate a target image on the basis of pupil coordinate data acquired from a learning image and learn weights of a plurality of filters in order to generate a filtered image, by filtering the learning image using the plurality of filters, that is within a predetermined reference range of the target image; a split image generator configured to generate a pupil region split image for an input image using the plurality of filters having the learned weights; and a pupil coordinate detector configured to remove noise of the pupil region split image, select at least one of a plurality of pixels from which noise is removed, and detect pupil coordinates.

Abstract: A fastening tool system for fastening a component part by using a supplied power includes: a camera portion mounted on a fastening tool to photograph a fastening portion of the component part; an image input portion for inputting an image photographed from the camera portion when a distance with the fastening portion measured through a distance measurement sensor is within a predetermined distance; a communication portion for transmitting data by wireless communication; and a fastening portion determining portion that recognizes the fastening portion through machine learning algorithm-based image classification work for an input image of the camera portion and sets a torque value matched with the recognized fastening portion.

Abstract: A video encoding/decoding apparatus according to the present invention acquires motion vector refinement information, performs motion compensation on the basis of a motion vector of a current block, refines the motion vector of the current block using at least one or both of the motion vector refinement information and the output of the motion compensation, and performs motion compensation using the refined motion vector.

Abstract: A video encoding/decoding apparatus according to the present invention acquires motion vector refinement information, performs motion compensation on the basis of a motion vector of a current block, refines the motion vector of the current block using at least one or both of the motion vector refinement information and the output of the motion compensation, and performs motion compensation using the refined motion vector.

Abstract: A video encoding/decoding apparatus according to the present invention acquires motion vector refinement information, performs motion compensation on the basis of a motion vector of a current block, refines the motion vector of the current block using at least one or both of the motion vector refinement information and the output of the motion compensation, and performs motion compensation using the refined motion vector.