Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020238320 - METHOD AND DEVICE FOR GENERATING EMOTICON

Publication Number WO/2020/238320
Publication Date 03.12.2020
International Application No. PCT/CN2020/078587
International Filing Date 10.03.2020
IPC
G06K 9/00 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR 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
CPC
G06F 16/784
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
16Information retrieval; Database structures therefor; File system structures therefor
70of video data
78Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
783using metadata automatically derived from the content
7837using objects detected or recognised in the video content
784the detected or recognised objects being people
G06F 16/7867
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
16Information retrieval; Database structures therefor; File system structures therefor
70of video data
78Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
7867using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
G06K 9/00302
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
00221Acquiring or recognising human faces, facial parts, facial sketches, facial expressions
00302Facial expression recognition
Applicants
  • 北京字节跳动网络技术有限公司 BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD. [CN]/[CN]
Inventors
  • 陈日伟 CHEN, Riwei
Agents
  • 北京天达共和律师事务所 EAST & CONCORD PARTNERS
Priority Data
201910447997.827.05.2019CN
Publication Language Chinese (ZH)
Filing Language Chinese (ZH)
Designated States
Title
(EN) METHOD AND DEVICE FOR GENERATING EMOTICON
(FR) PROCÉDÉ ET DISPOSITIF DE GÉNÉRATION D'ÉMOTICÔNE
(ZH) 用于生成表情包的方法和装置
Abstract
(EN)
Provided are a method and device for generating an emoticon. The method comprises: extracting an image sequence from a figure video to be processed (201), the image sequence comprising a plurality of target images on which faces are displayed; performing facial expression identification on the human face displayed on each target image in the image sequence to obtain a corresponding identification result (202), the identification result comprising a facial expression label and a facial expression degree corresponding to the facial expression label; and based on the facial expression degree corresponding to the facial expression label in the identification result corresponding to each target image, extracting a video clip from the figure video, and taking the video clip as an emoticon (203). According to the method, based on the facial expression matching, a video clip can be extracted from a given figure video as an emoticon, thereby achieving the generation of an emoticon based on facial expression matching.
(FR)
La présente invention concerne un procédé et un dispositif de génération d'émoticône. Le procédé consiste à : extraire une séquence d'images d'une vidéo de figure à traiter (201), la séquence d'images comprenant une pluralité d'images cibles sur lesquelles des visages sont affichés ; effectuer une identification d'expression faciale sur le visage humain affiché sur chaque image cible dans la séquence d'images afin d'obtenir un résultat d'identification correspondant (202), le résultat d'identification comprenant une étiquette d'expression faciale et un degré d'expression faciale correspondant à l'étiquette d'expression faciale ; et sur la base du degré d'expression faciale correspondant à l'étiquette d'expression faciale dans le résultat d'identification correspondant à chaque image cible, extraire un clip vidéo de la vidéo de figure, et prendre la séquence vidéo en tant qu'émoticône (203). Selon le procédé, sur la base de la correspondance d'expression faciale, un clip vidéo peut être extrait d'une vidéo de figure donnée en tant qu'émoticône, ce qui permet d'obtenir la génération d'une émoticône sur la base d'une correspondance d'expression faciale.
(ZH)
一种用于生成表情包的方法和装置,所述方法包括:从待处理的人物视频中提取出图像序列(201),其中,图像序列包括多张显示有人脸的目标图像;对图像序列中的各张目标图像分别显示的人脸进行表情识别,得到相应的识别结果(202),其中,识别结果包括表情标签和与表情标签对应的表情程度;基于该各张目标图像所对应的识别结果中的与该表情标签对应的表情程度,从人物视频中提取视频片段,并将该视频片段作为表情包(203)。所述方法可以基于人脸表情匹配,从给定的人物视频中提取出视频片段作为表情包,实现了基于人脸表情匹配的表情包的生成。
Also published as
Latest bibliographic data on file with the International Bureau