Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020113886 - BEHAVIOR FEATURE EXTRACTION METHOD, SYSTEM AND APPARATUS BASED ON TIME-SPACE/FREQUENCY DOMAIN HYBRID LEARNING

Publication Number WO/2020/113886
Publication Date 11.06.2020
International Application No. PCT/CN2019/083357
International Filing Date 19.04.2019
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
G06K 9/46 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
36Image preprocessing, i.e. processing the image information without deciding about the identity of the image
46Extraction of features or characteristics of the image
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
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
CPC
G06K 9/00
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
G06K 9/46
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
36Image preprocessing, i.e. processing the image information without deciding about the identity of the image
46Extraction of features or characteristics of the image
G06N 3/04
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architectures, e.g. interconnection topology
G06N 3/08
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
Applicants
  • 中国科学院自动化研究所 INSTITUTE OF AUTOMATION, CHINESE ACADEMY OF SCIENCES [CN]/[CN]
Inventors
  • 胡古月 HU, Guyue
  • 崔波 CUI, Bo
  • 余山 YU, Shan
Agents
  • 北京市恒有知识产权代理事务所(普通合伙) HENYOL INTELLECTUAL PROPERTY LAW CORPORATION
Priority Data
201811494799.907.12.2018CN
Publication Language Chinese (ZH)
Filing Language Chinese (ZH)
Designated States
Title
(EN) BEHAVIOR FEATURE EXTRACTION METHOD, SYSTEM AND APPARATUS BASED ON TIME-SPACE/FREQUENCY DOMAIN HYBRID LEARNING
(FR) PROCÉDÉ, SYSTÈME ET APPAREIL D'EXTRACTION DE CARACTÉRISTIQUES DE COMPORTEMENT BASÉS SUR UN APPRENTISSAGE HYBRIDE EN DOMAINES TEMPOREL/FRÉQUENTIEL
(ZH) 基于时空频域混合学习的行为特征提取方法、系统、装置
Abstract
(EN)
Disclosed are a behavior feature extraction method, system and apparatus based on time-space/frequency domain hybrid learning. The method comprises: acquiring a video behavior sequence based on a skeleton, and extracting a time-space domain behavior feature map by means of a conversion network; inputting same into a frequency domain attention network to perform frequency selection, then inverting same back to a time-space domain, and adding same to the time-space domain behavior feature map; synchronously performing local reasoning and non-local reasoning, and performing high-layer local reasoning; and globally pooling the time-space domain behavior feature map obtained by reasoning so as to obtain a behavior feature vector of the video behavior sequence.
(FR)
L'invention concerne un procédé, système et appareil d'extraction de caractéristiques de comportement basés sur un apprentissage hybride en domaines temporel/fréquentiel. Le procédé comporte les étapes consistant à: acquérir une séquence de comportement vidéo basée sur un squelette, et extraire une carte de caractéristiques de comportement en domaine spatio-temporel au moyen d'un réseau de conversion; introduire celle-ci dans un réseau d'attention en domaine fréquentiel pour effectuer une sélection de fréquences, puis l'inverser pour la ramener dans un domaine spatio-temporel, et l'additionner à la carte de caractéristiques de comportement en domaine spatio-temporel; effectuer de façon synchrone un raisonnement local et un raisonnement non local, et effectuer un raisonnement local en couche supérieure; et regrouper globalement la carte de caractéristiques de comportement en domaine spatio-temporel obtenue par raisonnement de façon à obtenir un vecteur de caractéristiques de comportement de la séquence de comportement vidéo.
(ZH)
一种基于时空频域混合学习的行为特征提取方法、系统、装置,方法包括:获取基于骨架的视频行为序列,通过变换网络提取时空域行为特征图;输入频域注意网络进行频率选择后逆变回时空域,与时空域行为特征图相加;同步进行局部和非局部推理,并进行高层局部推理;将推理得到的时空域行为特征图全局池化,得到视频行为序列的行为特征向量。
Also published as
Latest bibliographic data on file with the International Bureau