Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Human body action recognition method and device based on circular attention network

A human action recognition, cyclic neural network technology, applied in character and pattern recognition, biological neural network models, instruments, etc., to achieve the effect of good generalization and suppression of background noise

Pending Publication Date: 2020-05-26
NANJING NORMAL UNIVERSITY
View PDF5 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method has certain limitations and can only handle weakly labeled data containing one action type

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
  • Human body action recognition method and device based on circular attention network
  • Human body action recognition method and device based on circular attention network
  • Human body action recognition method and device based on circular attention network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings.

[0028] The present invention provides a method for automatic positioning, recognition and cutting of human motion sensor data based on a cyclic attention network. The overall flow chart of the algorithm is as follows figure 1 shown, including the following steps:

[0029] Step S1, in the case of third-party supervision and recording, collect the acceleration sensor data of the smart terminal device attached to the right wrist of the human body, and use it as a sample when training the human action recognition model.

[0030] Step S2, the sensor data is processed as follows: the sensor data is processed into weakly labeled data, that is, long-term sequence segments containing multiple action categories, and sequence tags are attached to the sequence segments. The final data format is (n, m, L , d), where n is the number of data, m is the numbe...

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 human body action recognition method and device based on a circular attention network. The method comprises the following steps: S1, collecting various action data of a humanbody by utilizing a multi-axis sensor; S2, segmenting the data into long-time sequence segments containing a plurality of action categories, and attaching sequence tags to the sequence segments; andS3, inputting the processed data into a pre-constructed multi-layer recurrent attention neural network model to realize recognition of specific types of actions. According to the invention, the cyclicattention algorithm is utilized, automatic positioning recognition and cutting of the sensor data can be achieved, and manpower and material resources consumed by manual labeling of the sensor data can be greatly reduced.

Description

technical field [0001] The invention belongs to the field of intelligent monitoring and recognition, and in particular relates to an automatic positioning recognition and cutting method based on human motion sensor data and a human body motion recognition device. Background technique [0002] Human action recognition technology has a wide range of needs in monitoring, human-computer interaction, assistive technology, sign language, computational behavioral science and consumer behavior analysis, monitoring, identification and analysis of human action behavior. In general, motion recognition technology is divided into two types: image-based recognition and sensor-based recognition, and sensor-based recognition is widely favored by researchers due to the convenience of data collection and the protection of user privacy. Most of the traditional methods for human action recognition belong to the scope of supervised learning. For example, in the early days, Support Vector Machin...

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): G06K9/00G06N3/04
CPCG06V40/20G06N3/045G06F2218/12
Inventor 张雷王焜严佳欢唐寅刘天一高文彬
Owner NANJING NORMAL UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com