Abstract: Provided is an image processing apparatus including a memory storing one or more instructions, and a processor executing the one or more instructions stored in the memory, wherein the processor retargets a current frame image, based on a first scale factor for a previous frame image, a second scale factor for the current frame image, and a weight determined based on previous frame image data and current frame image data.

Abstract: Disclosed is an image processing apparatus. The present image processing apparatus comprises: an input unit for inputting an image; and a processor for shrinking the inputted image to a predetermined ratio, extracting a visual feature from the shrunken image, performing an image quality enhancement process reflecting the extracted visual feature in the inputted image, repeatedly performing, for a predetermined number of times, the shrinking, the extracting, and the image quality enhancement process on the image that has undergone the image quality enhancement process. The present disclosure relates to an artificial intelligence (AI) system and an application thereof that simulate the functions of a human brain, such as recognition, judgment, etc., by using a machine learning algorithm such as deep learning, etc.

Abstract: An image processing device is disclosed. The image processing device includes: a memory storing one or more instructions; and a processor configured to execute the one or more instructions stored in the memory, wherein the processor is further configured to obtain a first image and a classifier that indicates a purpose of image processing, and to process, by using a deep neural network (DNN), the first image according to the purpose indicated by the classifier, wherein the DNN processes input images according to different purposes.

Abstract: Disclosed are a coded video data processing method and apparatus which consider a random access, and a coded video data generating method and apparatus which consider a random access. The coded video data processing method includes obtaining a bitstream of coded video data, obtaining metadata information used for video-processing of pictures having a decoding order after a random access point picture in the bitstream, and performing video-processing on decoded video data among the pictures having the decoding order after the random access point picture, based on the metadata information.

Abstract: Methods and apparatuses for encoding and decoding an intra prediction mode of a prediction unit of a chrominance component based on an intra prediction mode of a prediction unit of a luminance component are provided. When an intra prediction mode of a prediction unit of a luminance component is the same as an intra prediction mode in an intra prediction mode candidate group of a prediction unit of a chrominance component, reconstructing the intra prediction mode candidate group of the prediction unit of the chrominance component by excluding or replacing an intra prediction mode of the prediction unit of the chrominance component which is same as an intra prediction mode of the prediction unit of the luminance component from the intra prediction mode candidate group, and encoding the intra prediction mode of the prediction unit of the chrominance component by using the reconstructed intra prediction mode candidate group.

Abstract: Provided is an image processing apparatus for generating a high-resolution image. The image processing apparatus includes a memory storing one or more instructions and a processor configured to execute the one or more instructions stored in the memory, wherein the processor is configured to extract feature information regarding a low-resolution image of a current frame by using a first convolutional neural network, generate, based on the feature information, a first high-resolution image of the current frame, remove flickering of the first high-resolution image by using a high-resolution image of a previous frame, and remove flickering of a high-resolution image of a next frame by using at least one of a flickering-removed second high-resolution image of the current frame, or the feature information.

Abstract: Methods and apparatuses for encoding and decoding an intra prediction mode of a prediction unit of a chrominance component based on an intra prediction mode of a prediction unit of a luminance component are provided. When an intra prediction mode of a prediction unit of a luminance component is the same as an intra prediction mode in an intra prediction mode candidate group of a prediction unit of a chrominance component, reconstructing the intra prediction mode candidate group of the prediction unit of the chrominance component by excluding or replacing an intra prediction mode of the prediction unit of the chrominance component which is same as an intra prediction mode of the prediction unit of the luminance component from the intra prediction mode candidate group, and encoding the intra prediction mode of the prediction unit of the chrominance component by using the reconstructed intra prediction mode candidate group.

Abstract: Provided are methods and apparatus for encoding and decoding motion information. The method of encoding motion information includes: obtaining a motion information candidate by using motion information of prediction units that are temporally or spatially related to a current prediction unit; adding, when the number of motion information included in the motion information candidate is smaller than a predetermined number n, alternative motion information to the motion information candidate so that the number of motion information included in the motion information candidate reaches the predetermined number n; determining motion information with respect to the current prediction unit from among the n motion information candidates; and encoding index information indicating the determined motion information as motion information of the current prediction unit.

Abstract: Provided is an image processing apparatus including a memory storing one or more instructions and a processor configured to execute the one or more instructions stored in the memory, wherein the processor is further configured to execute the one or more instructions to generate a second image by performing a deconvolution operation on a first image and a kernel comprising one or more weights, set values of the one or more weights based on the second image, and adjust the values of the one or more weights based on positions of the one or more weights in the kernel.

Abstract: An image processing apparatus is provided. The image processing apparatus includes a processor configured to, in response to an image including a plurality of frames being input, change a predetermined parameter to a parameter corresponding to a compression rate of each of the plurality of frames for each frame, and process the input image by using the parameter changed for each frame, and an output interface configured to output the processed image.

Abstract: Provided are methods and apparatus for encoding and decoding motion information. The method of encoding motion information includes: obtaining a motion information candidate by using motion information of prediction units that are temporally or spatially related to a current prediction unit; adding, when the number of motion information included in the motion information candidate is smaller than a predetermined number n, alternative motion information to the motion information candidate so that the number of motion information included in the motion information candidate reaches the predetermined number n; determining motion information with respect to the current prediction unit from among the n motion information candidates; and encoding index information indicating the determined motion information as motion information of the current prediction unit.

Abstract: An image processing apparatus is disclosed. The present image processing apparatus comprises: a memory for storing a low dynamic range (LDR) image and a processor for adjusting the brightness of the LDR image by means of a pixel-specific brightness ratio identified using a first parameter, and acquiring a high dynamic range (HDR) image by adding or subtracting a pixel-specific correction value identified using a second parameter in the brightness-adjusted LDR image.

Abstract: An image processing apparatus is provided. The image processing apparatus according to an exemplary embodiment includes a communicator configured to receive an image, and a processor configured to generate a first image obtained by performing image processing on the received image by using a parameter for image processing, generate a second image obtained by reducing the first image at a predetermined ratio, and extract respective visual features from the first image and the second image, wherein the processor is further configured to adjust the parameter to allow a difference between the visual feature of the first image and the visual feature of the second image to be within a predetermined range.

Abstract: An image processing method of a display apparatus is disclosed. The image processing method includes analyzing a plurality of image frames constituting a virtual reality image and determining shake information of a camera which photographs the VR image, calculating a shake correction value of an image corresponding to a viewing area of a user from the VR image based on the shake information of the camera, and adjusting the viewing area based on the shake correction value and displaying an image corresponding to the adjusted viewing area in the VR image.

Abstract: A method and apparatus for intra predicting a video. The method includes: determining availability of a predetermined number of adjacent pixels used for intra prediction of a current block; if a first adjacent pixel is unavailable, searching for a second adjacent pixel that is available by searching the predetermined number of adjacent pixels in a predetermined direction based on the first adjacent pixel; and replacing a pixel value of the first adjacent pixel with a pixel value of a found second adjacent pixel. At least one third adjacent pixel in another location, which is not available and excludes the first adjacent pixel at a predetermined location, is to sequentially replaced by using a directly adjacent pixel in a predetermined direction.

Abstract: An electronic device is disclosed. The electronic device of the disclosure comprises: a memory in which a learned artificial intelligence model is stored; and a processor for inputting an input image to the artificial intelligence model and outputting an enlarged image with increased resolution, wherein the learned artificial intelligence model includes an upscaling module for acquiring the pixel values of interpolated pixels around a cell according to a function having a nonlinearly decreasing symmetric form with reference to an original pixel in the enlarged image, the original pixel corresponding to a pixel of the input image.

Abstract: The present disclosure relates to an artificial intelligence (AI) system utilizing a machine learning algorithm, including deep learning and the like, and application thereof. In particular, an electronic device of the present disclosure comprises: a memory including at least one command; and a processor connected to the memory so as to control the electronic device, wherein, by executing the at least one command, the processor acquires an image, acquires a noise correction map for correction of noise of the image on the basis of configuration information of a camera having captured the image or brightness information of the image, and eliminates the noise of the image through the noise correction map. In particular, at least a part of an image processing method may use an artificial intelligence model having been acquired through learning according to at least one of a machine learning algorithm, a neural network algorithm, and a deep learning algorithm.

Abstract: An image processing apparatus applies an image to a first learning network model to optimize the edges of the image, applies the image to a second learning network model to optimize the texture of the image, and applies a first weight to the first image and a second weight to the second image based on information on the edge areas and the texture areas of the image to acquire an output image.

Abstract: A method of performing, by an electronic device, a convolution operation at a certain layer in a neural network includes: obtaining N pieces of input channel data; performing a first convolution operation by applying a first input channel data group including K pieces of first input channel data from among the N pieces of input channel data to a first kernel filter group including K first kernel filters; performing a second convolution operation by applying a second input channel data group including K pieces of second input channel data from among the N pieces of input channel data to a second kernel filter group including K second kernel filters; and obtaining output channel data based on the first convolution operation and the second convolution operation, wherein K is a natural number that is less than N.

Abstract: A video encoding method and apparatus and a video decoding method and apparatus are provided. The video encoding method includes: prediction encoding in units of a coding unit as a data unit for encoding a picture, by using partitions determined based on a first partition mode and a partition level, so as to select a partition for outputting an encoding result from among the determined partitions; and encoding and outputting partition information representing a first partition mode and a partition level of the selected partition. The first partition mode represents a shape and directionality of a partition as a data unit for performing the prediction encoding on the coding unit, and the partition level represents a degree to which the coding unit is split into partitions for detailed motion prediction.