Processing

Please wait...

Settings

Settings

Goto Application

1. WO2021002719 - VIDEO PROCESSING METHOD AND APPARATUS

Publication Number WO/2021/002719
Publication Date 07.01.2021
International Application No. PCT/KR2020/008704
International Filing Date 03.07.2020
IPC
G06T 9/00 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
9Image coding
G06T 5/00 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
5Image enhancement or restoration
G06T 3/40 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
3Geometric image transformation in the plane of the image
40Scaling of a whole image or part thereof
G06N 3/08 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
G06N 3/04 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architecture, e.g. interconnection topology
Applicants
  • 한국과학기술원 KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY [KR]/[KR]
Inventors
  • 김문철 KIM, Mun Churl
  • 기세환 KI, Se Hwan
Agents
  • 특허법인 무한 MUHANN PATENT & LAW FIRM
Priority Data
10-2019-008036303.07.2019KR
10-2020-008091401.07.2020KR
Publication Language Korean (KO)
Filing Language Korean (KO)
Designated States
Title
(EN) VIDEO PROCESSING METHOD AND APPARATUS
(FR) PROCÉDÉ ET APPAREIL DE TRAITEMENT VIDÉO
(KO) 비디오 처리 방법 및 장치
Abstract
(EN)
A video processing method and apparatus are disclosed. A video processing method according to one embodiment comprises the steps of: receiving a video including a plurality of temporal portions; receiving a first model parameter corresponding to a first neural network for processing the entire video; receiving residues between the first model parameter and each of a plurality of second model parameters corresponding to a plurality of second neural networks for respectively processing the plurality of temporal portions; and performing, on the video, at least one from among super resolution, inverse tone mapping, tone mapping, frame interpolation, motion blur removal, de-noising, and compression distortion removal on the basis of the residues.
(FR)
L'invention concerne un procédé et un appareil de traitement vidéo. Un procédé de traitement vidéo selon un mode de réalisation comprend les étapes consistant à : recevoir une vidéo comprenant une pluralité de parties temporelles ; recevoir un premier paramètre de modèle correspondant à un premier réseau neuronal pour traiter l'ensemble de la vidéo ; recevoir des résidus entre le premier paramètre de modèle et chacun d'une pluralité de seconds paramètres de modèle correspondant à une pluralité de seconds réseaux neuronaux pour traiter respectivement la pluralité de parties temporelles ; et mettre en œuvre, sur la vidéo, au moins l'un parmi la super-résolution, le mappage de ton inverse, le mappage de ton, l'interpolation de trame, la suppression de flou de mouvement, le débruitage et l'élimination de distorsion de compression sur la base des résidus.
(KO)
비디오 처리 방법 및 장치가 개시된다. 일 실시예에 따른 비디오 처리 방법은, 복수의 시간적 부분(temporal portions)들을 포함하는 비디오를 수신하는 단계와, 상기 비디오 전체를 처리하기 위한 제1 뉴럴 네트워크에 대응하는 제1 모델 파라미터를 수신하는 단계와, 상기 복수의 시간적 부분들 각각을 처리하기 위한 복수의 제2 뉴럴 네트워크들에 대응하는 복수의 제2 모델 파라미터들 각각과 상기 제1 모델 파라미터 간의 잔차들(residues)을 수신하는 단계와, 상기 잔차들(residues)에 기초하여 상기 비디오에 초해상화, 역톤매핑, 톤매핑, 프레임 보간, 움직임 흐림 제거, 디노이징 및 압축 왜곡 제거 중에서 적어도 하나를 수행하는 단계를 포함한다.
Latest bibliographic data on file with the International Bureau