Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020093914 - CONTENT-WEIGHTED DEEP RESIDUAL LEARNING FOR VIDEO IN-LOOP FILTERING

Publication Number WO/2020/093914
Publication Date 14.05.2020
International Application No. PCT/CN2019/114259
International Filing Date 30.10.2019
IPC
H04N 19/00 2014.01
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
NPICTORIAL COMMUNICATION, e.g. TELEVISION
19Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
CPC
G06K 9/6228
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
6228Selecting the most significant subset of features
G06K 9/6261
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
6261partitioning the feature space
H04N 19/82
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
NPICTORIAL COMMUNICATION, e.g. TELEVISION
19Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
80Details of filtering operations specially adapted for video compression, e.g. for pixel interpolation
82involving filtering within a prediction loop
H04N 19/86
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
NPICTORIAL COMMUNICATION, e.g. TELEVISION
19Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
85using pre-processing or post-processing specially adapted for video compression
86involving reduction of coding artifacts, e.g. of blockiness
Applicants
  • ALIBABA GROUP HOLDING LIMITED
Inventors
  • DU, Yixin
  • LOU, Jian
  • SUN, Yuchen
Agents
  • BEIJING SANYOU INTELLECTUAL PROPERTY AGENCY LTD.
Priority Data
16/184,84208.11.2018US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) CONTENT-WEIGHTED DEEP RESIDUAL LEARNING FOR VIDEO IN-LOOP FILTERING
(FR) APPRENTISSAGE RÉSIDUEL PROFOND PONDÉRÉ PAR UN CONTENU DESTINÉ À UN FILTRAGE EN BOUCLE VIDÉO
Abstract
(EN)
Systems and methods are provided for improving filtering performance and Bjøntegaard-Delta (BD) rate savings for video processing. In addition to computing the artifacts between a given compressed image and a restored clean image after filtering using Deep Residual Learning (DRL) for recovering the residual between input and output, filtering strength of a loop filter may be controlled by the content of the region of the image, such that, in more important areas, such as the face and edges, the filtering strength may be increased while in less important areas, such as textures and backgrounds, the filtering strength may be decreased.
(FR)
La présente invention concerne des systèmes et des procédés destinés à améliorer les performances de filtrage et les économies de taux de Bjøntegaard-Delta (BD) pour un traitement vidéo. En plus de calculer les artéfacts entre une image compressée donnée et une image propre restaurée après filtrage à l'aide d'un apprentissage résiduel profond (DRL) pour récupérer le résidu entre l'entrée et la sortie, la force de filtrage d'un filtre à boucle peut être commandée par le contenu de la région de l'image, de telle sorte que, dans des zones plus importantes, telles que la face et les bords, la force de filtrage peut être augmentée alors qu'elle se trouve dans des zones moins importantes, telles que des textures et des fonds, la force de filtrage peut être réduite.
Also published as
Latest bibliographic data on file with the International Bureau