Abstract: An image processing device includes an encoder which infers latent variables from an input noisy image based on a preset noise. A denoiser removes the noise from the input noisy image to generate a denoising image. A decoder reconstructs a noisy image, using the inferred latent variable. The latent variables are learned on the basis of a difference between the reconstructed noisy image and the input noisy image.

Abstract: Disclosed is a method for photographing a panoramic image including the steps of recognizing movement of a corresponding photographing apparatus by comparing a current real-time input image with a previous image through a motion estimation mechanism with exposure compensation, determining a time to photograph each next picture by determining whether movement in a photography direction reaches a preset threshold value, and photographing each next picture by manual or automatic operation at the determined time.

Abstract: A device for generating a high-resolution frame includes a plurality of alignment circuits configured to generate a plurality of aligned frames by blending a reference frame and a plurality of neighboring frames neighboring the reference frame; and a reconstruction circuit configured to generate the high-resolution frame corresponding to the reference frame according to the reference frame and the plurality of aligned frames. The plurality of alignment circuits and the reconstruction circuit each include neural networks.

Abstract: The disclosure relates to a method of determining a gaze distance, the method including obtaining a plurality of images by capturing each of both eyes of a user, determining, from each of the plurality of images, a corner point located at an edge of the eye, a location of a pupil, and an eye contour, determining each of gaze directions of both of the eyes based on a difference between a reference point and the location of the pupil, wherein the reference point is determined based on the eye contour and a location of the corner point, and determining a gaze distance based on a difference between the gaze directions of both of the eyes.

Abstract: An electronic apparatus is provided. The electronic apparatus includes a storage configured to store a compression rate network model configured to determine a compression rate applied to an image block from among a plurality of compression rates, and a plurality of compression noise removing network models configured to remove compression noise for each of the plurality of compression rates, and a processor configured to: obtain a compression rate of each of a plurality of image blocks included in a frame of a decoded moving picture based on the compression rate network model, obtain the compression rate of the frame based on the plurality of obtained compression rates, and remove compression noise of the frame based on a compression noise removing network model corresponding to the compression rate of the frame from among the plurality of compression noise removing network models.

Abstract: Disclosed is an image processing apparatus and method for generating one HDR image by using a plurality of LDR images photographed at different exposure values. The image processing apparatus may include a deep learning framework configured to receive a plurality of low dynamic range (LDR) images captured at different exposures, generate kernels for alignment by using the plurality of LDR images, generate aligned images by applying the kernels to the plurality of LDR images, and generate a high dynamic range (HDR) image by synthesizing the aligned images.

Abstract: The disclosure relates to a method of determining a gaze distance, the method including obtaining a plurality of images by capturing each of both eyes of a user, determining, from each of the plurality of images, a corner point located at an edge of the eye, a location of a pupil, and an eye contour, determining each of gaze directions of both of the eyes based on a difference between a reference point and the location of the pupil, wherein the reference point is determined based on the eye contour and a location of the corner point, and determining a gaze distance based on a difference between the gaze directions of both of the eyes.

Abstract: Disclosed is a method for photographing a panoramic image including the steps of recognizing movement of a corresponding photographing apparatus by comparing a current real-time input image with a previous image through a motion estimation mechanism with exposure compensation, determining a time to photograph each next picture by determining whether movement in a photography direction reaches a preset threshold value, and photographing each next picture by manual or automatic operation at the determined time.

Abstract: Disclosed is a method for photographing a panoramic image including the steps of recognizing movement of a corresponding photographing apparatus by comparing a current real-time input image with a previous image through a motion estimation mechanism with exposure compensation, determining a time to photograph each next picture by determining whether movement in a photography direction reaches a preset threshold value, and photographing each next picture by manual or automatic operation at the determined time.

Abstract: An electronic apparatus is provided. The electronic apparatus includes a storage configured to store a compression rate network model configured to determine a compression rate applied to an image block from among a plurality of compression rates, and a plurality of compression noise removing network models configured to remove compression noise for each of the plurality of compression rates, and a processor configured to: obtain a compression rate of each of a plurality of image blocks included in a frame of a decoded moving picture based on the compression rate network model, obtain the compression rate of the frame based on the plurality of obtained compression rates, and remove compression noise of the frame based on a compression noise removing network model corresponding to the compression rate of the frame from among the plurality of compression noise removing network models.

Abstract: Disclosed is a method for photographing a panoramic image including the steps of recognizing movement of a corresponding photographing apparatus by comparing a current real-time input image with a previous image through a motion estimation mechanism with exposure compensation, determining a time to photograph each next picture by determining whether movement in a photography direction reaches a preset threshold value, and photographing each next picture by manual or automatic operation at the determined time.

Abstract: Disclosed is a method for photographing a panoramic image including the steps of recognizing movement of a corresponding photographing apparatus by comparing a current real-time input image with a previous image through a motion estimation mechanism with exposure compensation, determining a time to photograph each next picture by determining whether movement in a photography direction reaches a preset threshold value, and photographing each next picture by manual or automatic operation at the determined time.

Abstract: The present disclosure relates an image correction apparatus and method. The image correction apparatus includes a printing unit for printing an image; a scratch detecting unit which detects a scratch region from a scan image scanned per line; and a restoration processing unit which performs the restoration of the scratch region by interpolation if the width of the scratch region detected by the scratch detecting unit is less than a predetermined value and, if the width thereof is not less than the predetermined value, does not perform a restoration of the scratch region but provides the printing unit with the scan image, thereby correcting an image distortion by easily detecting an image distortion due to foreign matter.

Abstract: Disclosed is a method for photographing a panoramic image including the steps of recognizing movement of a corresponding photographing apparatus by comparing a current real-time input image with a previous image through a motion estimation mechanism with exposure compensation, determining a time to photograph each next picture by determining whether movement in a photography direction reaches a preset threshold value, and photographing each next picture by manual or automatic operation at the determined time.

Abstract: Disclosed is a method for photographing a panoramic image including the steps of recognizing movement of a corresponding photographing apparatus by comparing a current real-time input image with a previous image through a motion estimation mechanism with exposure compensation, determining a time to photograph each next picture by determining whether movement in a photography direction reaches a preset threshold value, and photographing each next picture by manual or automatic operation at the determined time.

Abstract: A method and apparatus for generating a dewarped document using a document image captured using a camera are provided. The method includes obtaining the document image captures using the camera, extracting text lines from the document image captured using the camera, determining a projection formula to convert positions of respective points constituting the extracted text lines to coordinates projected on a plane of the dewarped document, determining a target function used to calculate a difference between text lines projected on the place of the dewarped document using the projection formula and real text lines, calculating parameters that minimize the target function, and converting the document image to the dewarped document by substituting the calculated parameters into the projection formula.

Abstract: The present disclosure relates an image correction apparatus and method. The image correction apparatus includes a printing unit for printing an image; a scratch detecting unit which detects a scratch region from a scan image scanned per line; and a restoration processing unit which performs the restoration of the scratch region by interpolation if the width of the scratch region detected by the scratch detecting unit is less than a predetermined value and, if the width thereof is not less than the predetermined value, does not perform a restoration of the scratch region but provides the printing unit with the scan image, thereby correcting an image distortion by easily detecting an image distortion due to foreign matter.

Abstract: A method of recognizing and generating a structure of a table included in an image is provided. The method includes extracting lines forming the table from among connection components forming an image, determining line intersections by using crossing functions matched with the lines, determining one of a plurality of crossing models identified based on a plurality of crossing shapes, in correspondence with each of the line intersections, and generating data about the table, which includes at least one cell determined using the determined crossing model.

Abstract: A method and apparatus for generating a dewarped document using a document image captured using a camera are provided. The method includes obtaining the document image captures using the camera, extracting text lines from the document image captured using the camera, determining a projection formula to convert positions of respective points constituting the extracted text lines to coordinates projected on a plane of the dewarped document, determining a target function used to calculate a difference between text lines projected on the place of the dewarped document using the projection formula and real text lines, calculating parameters that minimize the target function, and converting the document image to the dewarped document by substituting the calculated parameters into the projection formula.

Abstract: A method of recognizing and generating a structure of a table included in an image is provided. The method includes extracting lines forming the table from among connection components forming an image, determining line intersections by using crossing functions matched with the lines, determining one of a plurality of crossing models identified based on a plurality of crossing shapes, in correspondence with each of the line intersections, and generating data about the table, which includes at least one cell determined using the determined crossing model.