公开(公告)号：US20230114954A1

公开(公告)日：2023-04-13

申请号：US17887236

申请日：2022-08-12

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Hyunseung LEE ,  Donghyun KIM ,  Younghoon JEONG

IPC: G06T3/40 ,  G06V10/25

Abstract: A display device for performing image processing by using a neural network including a plurality of layers, may obtain a plurality of pieces of model information respectively corresponding to pixels included in a first image based on object features respectively corresponding to the pixels; identify the plurality of pieces of model information respectively corresponding to the plurality of layers and the pixels input to the neural network based on information about a time point at which each of the pixels is processed in the neural network; update parameters of the plurality of layers, based on the plurality of pieces of model information; and obtain a second image by processing the first image via the plurality of layers to which the updated parameters are applied; and display the second image.

公开(公告)号：US20190281310A1

公开(公告)日：2019-09-12

申请号：US16292655

申请日：2019-03-05

Applicant: SAMSUNG ELECTRONICS CO., LTD. ,  SEOUL NATIONAL UNIVERSITY R&DB FOUNDATION

Inventor： Hyunseung LEE ,  Donghyun KIM ,  Youngsu MOON ,  Taegyoung AHN ,  Yoonsik KIM ,  Jaewoo PARK ,  Jae Woong SOH ,  Nam Ik CHO ,  Byeongyong AHN

IPC: H04N19/176 ,  H04N19/137 ,  G06N3/02 ,  G06K9/40

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. The compression rate network model can be obtained by learning image characteristics of a plurality of restored image blocks corresponding to each of the plurality of compression rates through a first artificial intelligence algorithm, and the plurality of restored image blocks can be generated by encoding a plurality of original image blocks, and decoding the encoded plurality of original image blocks, and the plurality of compression noise removing network models can be obtained by learning a relation between the plurality of original image blocks and the plurality of restored image blocks through a second artificial intelligence algorithm.

公开(公告)号：US20200372608A1

公开(公告)日：2020-11-26

申请号：US16838650

申请日：2020-04-02

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Cheon LEE ,  Donghyun KIM ,  Yongsup PARK ,  Jaeyeon PARK ,  Iljun AHN ,  Hyunseung LEE ,  Taegyoung AHN ,  Youngsu MOON ,  Tammy LEE

IPC: G06T3/40 ,  G06T7/13 ,  G06N3/08

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.

公开(公告)号：US20200349677A1

公开(公告)日：2020-11-05

申请号：US16862958

申请日：2020-04-30

Applicant: SAMSUNG ELECTRONICS CO., LTD. ,  KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY

Inventor： Hyunseung LEE ,  MunChurl KIM ,  Yongwoo KIM ,  Jae Seok CHOI ,  Youngsu MOON ,  Cheon LEE

IPC: G06T5/00 ,  G06T5/50 ,  G06T7/13

Abstract: An image processing apparatus obtains a first output image by applying an image to a first training network model, obtains a second output image by applying the image to a second training network model, and obtains a reconstructed image based on the first output image and the second output image. The first training network model is a model that uses a fixed parameter obtained through training of a plurality of sample images, the second training network model is trained to minimize a difference between a target image corresponding to the image and the reconstructed image.

公开(公告)号：US20240412325A1

公开(公告)日：2024-12-12

申请号：US18811938

申请日：2024-08-22

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Hyunseung LEE ,  Donghyun KIM ,  Youngsu MOON ,  Seungho PARK ,  Younghoon JEONG

IPC: G06T3/4046 ,  G06F18/21 ,  G06N3/045 ,  G06N3/08

Abstract: An electronic apparatus is disclosed. The electronic apparatus includes a memory configured to store a plurality of neural network models, and a processor connected to the memory and control the electronic apparatus in which the processor is configured to obtain a weight map based on an object area included in an input image, and obtain a plurality of images by inputting the input image to each of the plurality of neural network models, and obtain an output image by weighting the plurality of images based on the weight map, and each of the plurality of neural network models is a model trained to upscale an image.

公开(公告)号：US20230360383A1

公开(公告)日：2023-11-09

申请号：US18143326

申请日：2023-05-04

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Hyunseung LEE ,  Youngsu MOON

IPC: G06V10/82 ,  G06V10/72

CPC classification number: G06V10/82 ,  G06V10/72

Abstract: An image processing method including obtaining a meta model based on a quality of an input image, training the meta model by using a training data set corresponding to the input image, and obtaining a quality-processed output image from the input image based on the trained meta model.

公开(公告)号：US20240037377A1

公开(公告)日：2024-02-01

申请号：US18345533

申请日：2023-06-30

Applicant: Samsung Electronics Co., Ltd. ,  Seoul National University R&DB Foundation

Inventor： Hoon SHIN ,  Jae Wook LEE ,  Rihae PARK ,  Yeonhong PARK ,  Seung Yul LEE ,  Hyunseung LEE

IPC: G06N3/048 ,  G11C11/54

CPC classification number: G06N3/048 ,  G11C11/54

Abstract: A method and apparatus are provided. The method includes reordering a plurality of filters, then based on a result of the reordering, compressing weights, among a plurality of weights of the plurality of filters, resulting in some of the plurality of weights being uncompressed weights, generating a plurality of operation unit maps by mapping the uncompressed weights to respective operation units according to a predetermined bulk unit, and mapping the plurality of operation unit maps to an array.

公开(公告)号：US20220164923A1

公开(公告)日：2022-05-26

申请号：US17221105

申请日：2021-04-02

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Hyunseung LEE ,  Donghyun KIM ,  Youngsu MOON ,  Seungho PARK ,  Younghoon JEONG

IPC: G06T3/40 ,  G06N3/08 ,  G06N3/04 ,  G06K9/62

Abstract: An electronic apparatus is disclosed. The electronic apparatus includes a memory configured to store a plurality of neural network models, and a processor connected to the memory and control the electronic apparatus in which the processor is configured to obtain a weight map based on an object area included in an input image, and obtain a plurality of images by inputting the input image to each of the plurality of neural network models, and obtain an output image by weighting the plurality of images based on the weight map, and each of the plurality of neural network models is a model trained to upscale an image.

公开(公告)号：US20250045881A1

公开(公告)日：2025-02-06

申请号：US18923346

申请日：2024-10-22

Applicant: SAMSUNG ELECTRONICS CO., LTD. ,  SEOUL NATIONAL UNIVERSITY R&DB FOUNDATION

Inventor： Hyunseung LEE ,  Seokhee LEE ,  Kyuha CHOI ,  Youngsu MOON ,  Karam PARK ,  Younghoon JEONG ,  Namik CHO ,  Sunwoo CHO

IPC: G06T5/60 ,  G06T3/4046 ,  G06T3/4053 ,  G06T5/20

Abstract: An image processing apparatus, including: at least one processor; and a memory configured to store one or more instructions which, when executed by the at least one processor, cause the image processing apparatus to: obtain a neural network model for performing image quality processing on an input image; calculate a plurality of gradients by partially differentiating weights of the neural network model with respect to a loss of the neural network model, by applying training data corresponding to the input image to the neural network model; remove at least one gradient from among the plurality of gradients by applying a gradient mask including gradient pruning information to the plurality of gradients; train the neural network model by updating the weights of the neural network model based on one or more remaining gradients from among the plurality of gradients; and obtain a quality-processed output image based on the input image, using the trained neural network model

公开(公告)号：US20240078631A1

公开(公告)日：2024-03-07

申请号：US18506755

申请日：2023-11-10

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Cheon LEE ,  Donghyun KIM ,  Yongsup PARK ,  Jaeyeon PARK ,  Iljun AHN ,  Hyunseung LEE ,  Taegyoung AHN ,  Youngsu MOON ,  Tammy LEE

IPC: G06T3/40 ,  G06T7/13 ,  G06V10/82

CPC classification number: G06T3/403 ,  G06T3/4053 ,  G06T7/13 ,  G06V10/82 ,  G06T2207/20081

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.