公开(公告)号：US20230075881A1

公开(公告)日：2023-03-09

申请号：US17987582

申请日：2022-11-15

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jaeyeon PARK ,  Iljun AHN ,  Tammy LEE

IPC: G06T5/00 ,  G06T5/50 ,  G06T5/20 ,  G06T3/40

Abstract: Provided is a method of reducing noise of a compressed image, the method including obtaining a first image by applying a convolution filter, which sequentially performs down-convolution and up-convolution corresponding to the down-convolution, to a compressed image of an original image, obtaining a second image by subtracting the first image from the compressed image, obtaining noise comprising high-frequency information and a compressed artifact of the compressed image from the second image, obtaining a third image by removing the compressed artifact by applying a deep neural network (DNN) for removing the compressed artifact to the noise, and reconstructing the compressed image by summing the first image and the third image.

公开(公告)号：US20220246158A1

公开(公告)日：2022-08-04

申请号：US17722569

申请日：2022-04-18

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Woohyun NAM ,  Sangchul KO ,  Kyungrae KIM ,  Jungkyu KIM ,  Yoonjae SON ,  Tammy LEE ,  Hyunkwon CHUNG ,  Sunghee HWANG

IPC: G10L19/008 ,  H04S3/00 ,  G06N3/08

Abstract: An audio processing apparatus may obtain second audio signals corresponding to channels included in a second channel group from first audio signals corresponding to channels included in a first channel group, downsample at least one third audio signal corresponding to at least one channel identified based on a correlation with the second channel group from among the channels included in the first channel group, by using an artificial intelligence (AI) model, and generate a bitstream including the second audio signals corresponding to the channels included in the second channel group and the downsampled at least one third audio signal. The first channel group includes a channel group of an original audio signal, and the second channel group is constructed by combining at least two channels from among the channels included in the first channel group.

公开(公告)号：US20220067880A1

公开(公告)日：2022-03-03

申请号：US17413404

申请日：2019-10-10

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Kyungmin LIM ,  Jaesung LEE ,  Tammy LEE

IPC: G06T3/40 ,  G06T7/73 ,  G06N3/04

Abstract: An image processing apparatus includes a memory storing one or more instructions, and a processor executing the one or more instructions stored in the memory, wherein the processor uses one or more deep neural networks to generate first feature maps from a first image, generate second feature maps for each of subpixels based on the first feature maps and position information of each of the subpixels arranged in a display, and determine result values for each of the subpixels based on the second feature maps.

公开(公告)号：US20220019844A1

公开(公告)日：2022-01-20

申请号：US17299607

申请日：2019-10-16

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jaeyeon PARK ,  Iljun AHN ,  Yongsup PARK ,  Jaehee KWAK ,  Tammy LEE ,  Minsu CHEON

IPC: G06K9/62 ,  G06N3/04 ,  G06T7/11 ,  G06K9/46

Abstract: An 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. The processor is configured to, by using one or more convolution neural networks, extract target features by performing a convolution operation between features of target regions having same locations in a plurality of input images and a first kernel set, extract peripheral features by performing a convolution operation of features of peripheral regions located around the target regions in the plurality of input images and a second kernel set, and determine a feature of a region corresponding to the target regions in an output image, based on the target features and the peripheral features.

公开(公告)号：US20200084456A1

公开(公告)日：2020-03-12

申请号：US16680133

申请日：2019-11-11

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Tammy LEE ,  Jianle CHEN

IPC: H04N19/159 ,  H04N19/105 ,  H04N19/176 ,  H04N19/11 ,  H04N19/136 ,  H04N19/182 ,  H04N19/593

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.

公开(公告)号：US20200058113A1

公开(公告)日：2020-02-20

申请号：US16348584

申请日：2017-11-08

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Woo-hyun NAM ,  Il-jun AHN ,  Tammy LEE ,  Ki-heum CHO ,  Yong-sup PARK ,  Min-su CHEON

IPC: G06T5/50 ,  G06T3/40

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.

公开(公告)号：US20190272621A1

公开(公告)日：2019-09-05

申请号：US16345616

申请日：2017-10-26

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Il-jun AHN ,  Woo-hyun NAM ,  Ki-heum CHO ,  Yong-sup PARK ,  Tammy LEE ,  Min-su CHEON

IPC: G06T3/40 ,  G06T5/50 ,  G06T5/00

Abstract: An image processing apparatus is disclosed. The present image processing apparatus includes an input unit to which an image is input; and a processor which extracts visual characteristics by reducing an input image and obtains a high-definition image by reflecting extracted visual characteristics on the input image. The disclosure relates to an artificial intelligence (AI) system and application thereof that simulate functions such as cognition and decision-making of a human brain using a machine learning algorithm such as deep learning.

公开(公告)号：US20190073757A1

公开(公告)日：2019-03-07

申请号：US16121057

申请日：2018-09-04

Applicant: Samsung Electronics Co., Ltd.

Inventor： Il-jun AHN ,  Ki-heum CHO ,  Yong-sup PARK ,  Jae-yeon PARK ,  Hee-seok OH ,  Tammy LEE ,  Min-su CHEON

IPC: G06T5/00 ,  G06K9/00 ,  G06T5/50

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.

公开(公告)号：US20190028733A1

公开(公告)日：2019-01-24

申请号：US16137594

申请日：2018-09-21

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Tammy LEE ,  Jianle CHEN

IPC: H04N19/52 ,  H04N19/82 ,  H04N19/91 ,  H04N19/593 ,  H04N19/51 ,  H04N19/13

Abstract: Provided are a method and apparatus for intra predicting an image, which generate a prediction value via linear interpolation in horizontal and vertical directions of a current prediction unit. The method includes: generating first and second virtual pixels by using at least one adjacent pixel located upper right and lower left to a current prediction unit; obtaining a first prediction value of a current pixel via linear interpolation using an adjacent left pixel located on the same line as the first virtual pixel and the current pixel; obtaining a second prediction value of the current pixel via linear interpolation using an adjacent upper pixel located on the same column as the second virtual pixel and the current pixel; and obtaining a prediction value of the current pixel by using the first and second prediction values.

公开(公告)号：US20180310015A1

公开(公告)日：2018-10-25

申请号：US16016031

申请日：2018-06-22

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Woo-jin HAN ,  Elena ALSHINA ,  Jung-hye MIN ,  Alexander ALSHIN ,  Il-koo KIM ,  Nikolay SHLYAKHOV ,  Tammy LEE ,  Sun-il LEE ,  Min-su CHEON ,  Jianle CHEN ,  Vadim SEREGIN ,  Yoon-mi HONG

IPC: H04N19/50 ,  H04N19/96 ,  H04N19/86

CPC classification number: H04N19/59 ,  H04N19/117 ,  H04N19/119 ,  H04N19/139 ,  H04N19/147 ,  H04N19/159 ,  H04N19/172 ,  H04N19/176 ,  H04N19/184 ,  H04N19/197 ,  H04N19/30 ,  H04N19/44 ,  H04N19/50 ,  H04N19/513 ,  H04N19/52 ,  H04N19/593 ,  H04N19/61 ,  H04N19/80 ,  H04N19/82 ,  H04N19/86 ,  H04N19/96

Abstract: A method of encoding a video is provided, the method includes: determining a filtering boundary on which deblocking filtering is to be performed based on at least one data unit from among a plurality of coding units that are hierarchically configured according to depths indicating a number of times at least one maximum coding unit is spatially spilt, and a plurality of prediction units and a plurality of transformation units respectively for prediction and transformation of the plurality of coding units, determining filtering strength at the filtering boundary based on a prediction mode of a coding unit to which pixels adjacent to the filtering belong from among the plurality of coding units, and transformation coefficient values of the pixels adjacent to the filtering boundary, and performing deblocking filtering on the filtering boundary based on the determined filtering strength.