Traitement en cours

Veuillez attendre...

Paramétrages

Paramétrages

Aller à Demande

1. CN109558832 - Human body posture detection method, device and equipment and storage medium

Office Chine
Numéro de la demande 201811427578.X
Date de la demande 27.11.2018
Numéro de publication 109558832
Date de publication 02.04.2019
Type de publication A
CIB
G06K 9/00
GPHYSIQUE
06CALCUL; COMPTAGE
KRECONNAISSANCE DES DONNÉES; PRÉSENTATION DES DONNÉES; SUPPORTS D'ENREGISTREMENT; MANIPULATION DES SUPPORTS D'ENREGISTREMENT
9Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
G06K 9/62
GPHYSIQUE
06CALCUL; COMPTAGE
KRECONNAISSANCE DES DONNÉES; PRÉSENTATION DES DONNÉES; SUPPORTS D'ENREGISTREMENT; MANIPULATION DES SUPPORTS D'ENREGISTREMENT
9Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
62Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques
CPC
G06K 9/00342
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
00335Recognising movements or behaviour, e.g. recognition of gestures, dynamic facial expressions; Lip-reading
00342Recognition of whole body movements, e.g. for sport training
G06K 9/00369
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
00362Recognising human body or animal bodies, e.g. vehicle occupant, pedestrian; Recognising body parts, e.g. hand
00369Recognition of whole body, e.g. static pedestrian or occupant recognition
G06K 9/6256
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
6256Obtaining sets of training patterns; Bootstrap methods, e.g. bagging, boosting
Déposants GUANGZHOU BIGO INFORMATION TECHNOLOGY CO., LTD.
广州市百果园信息技术有限公司
Inventeurs XIANG WEI
项伟
WANG YIFENG
王毅峰
HUANG QIUSHI
黄秋实
LIANG ZHUJIN
梁柱锦
Mandataires 北京品源专利代理有限公司 11332
Titre
(EN) Human body posture detection method, device and equipment and storage medium
(ZH) 一种人体姿态检测方法、装置、设备及存储介质
Abrégé
(EN)
The invention discloses a human body posture detection method, device and equipment and a storage medium. The method comprises the steps of collecting multi-frame image data; inputting the current frame of image data into a pre-trained human body posture detection model, outputting a plurality of human body posture reference graphs by referring to the human body posture confidence graph of the previous frame of image data, and generating the human body posture detection model by training through a convolutional neural network applied to an embedded platform; identifying human body posture keypoints in the human body posture reference map; generating a human body posture confidence map according to the credibility of the human body posture key points; judging whether the current frame of image data is the last frame of image data or not; if not, inputting the human body posture confidence map into the human body posture detection model for participating in generating the human body posture confidence map of the next frame of image data; and if yes, ending the execution of the operation of generating the human body posture confidence map of the multi-frame image data. According to the embodiment of the invention, human body posture detection on the embedded platform is realized.

(ZH)
本发明公开了一种人体姿态检测方法、装置、设备及存储介质。该方法包括:采集多帧图像数据;将当前帧图像数据输入至预先训练的人体姿态检测模型中,以参考上一帧图像数据的人体姿态置信图,输出多张人体姿态参考图,人体姿态检测模型为经应用于嵌入式平台的卷积神经网络训练生成;在人体姿态参考图中识别人体姿态关键点;根据人体姿态关键点的可信性,生成人体姿态置信图;判断当前帧图像数据是否为最后一帧图像数据;若否,则将人体姿态置信图输入至人体姿态检测模型中,用于参与生成下一帧图像数据的人体姿态置信图;若是,则结束执行生成多帧图像数据的人体姿态置信图的操作。本发明实施例实现了在嵌入式平台上进行人体姿态检测。

Également publié en tant que