Real-time human body posture recognition method under complex environment based on bidirectional LSTM

A technology of human body posture and complex environment, which is applied in the field of human-computer interaction, and can solve problems that affect the accuracy of posture recognition and interference

Pending Publication Date: 2020-06-09
SHANGHAI UNIV
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But on the other hand, the features acquired by vision will cause great interference due to changes in the surrounding environment, which will significantly affect the accuracy of gesture recognition.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Real-time human body posture recognition method under complex environment based on bidirectional LSTM
  • Real-time human body posture recognition method under complex environment based on bidirectional LSTM
  • Real-time human body posture recognition method under complex environment based on bidirectional LSTM

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0077] Specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings.

[0078] This implementation uses the OpenPose framework to obtain two-dimensional key point information of the human body, and the specific display can be seen Figure 4 . In this embodiment, all data transmission relies on the ROS operating system, obtains the current RGB image and depth image through the Topic released by Kinect, and sends the obtained two-dimensional key point information through the service for other modules to call.

[0079] Such as figure 1 As shown, a real-time human gesture recognition method in a complex environment based on bidirectional LSTM includes the following steps:

[0080] Step 1: Obtain the two-dimensional key point coordinates of the human body P=(p 0 ,p 1 ,...,p l ), where l is the number of key points, p i =(u i ,v i );

[0081] Step 1.1: Use the Kinect visual sensor to acquire each frame of two-d...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a real-time human body posture recognition method in a complex environment based on bidirectional LSTM. OpenPose is used as a human body posture estimation module to obtain two-dimensional joint point data of a human body. And judging whether the human body is in a shielded state or not according to the data missing condition. For a non-occlusion condition, constructing a classifier based on bidirectional LSTM, and sending the initial two-dimensional joint point information to the classifier to obtain a human body posture of the non-occlusion condition; and for a shielding state, performing three-dimensional mapping by using internal parameters of a depth camera, constructing a trunk vector and a joint angle, processing the high-dimensional features by using principal component analysis, and sending the processed high-dimensional features to a classifier to obtain a human body posture of the shielding condition. According to the method, the human body posture can be accurately recognized in a complex environment.

Description

technical field [0001] The invention belongs to the field of human-computer interaction technology, and more specifically, relates to a real-time human body posture recognition method in a complex environment based on a bidirectional LSTM. Background technique [0002] The application of robot technology has gradually expanded from the traditional industrial field to scenarios with frequent interactions with people, such as medical care and services, because the above scenarios have extremely high requirements for safety and reliability. Human-computer interaction technology has made great progress. People need robots to quickly understand various information transmitted by humans, and make more natural decisions and feedback based on different information characteristics. [0003] On the one hand, human posture is the most intuitive way to represent the human state, and on the other hand, it has become an important research direction in human-computer interaction because i...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/08
CPCG06N3/08G06V40/10G06V40/20G06F18/24
Inventor 周意乔徐昱琳
Owner SHANGHAI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap