Unlock instant, AI-driven research and patent intelligence for your innovation.

Human body training action recognition and counting method and system

A counting method, human body technology, applied in the field of action recognition, can solve the problems of lack of supervision, difficult to supervise the movement effectively, etc.

Pending Publication Date: 2022-08-05
长三角信息智能创新研究院 +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Traditional methods, such as requiring students to consciously do endurance training after class, are difficult to effectively supervise the exercise, and require trainers to complete it consciously, lacking supervision

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 training action recognition and counting method and system
  • Human body training action recognition and counting method and system
  • Human body training action recognition and counting method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] A preferred embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

[0027] The identification and counting method of human body training action in the present invention comprises the following steps:

[0028] S1: Collect the video of human training, and extract the skeleton point data of the human body in the video through a lightweight convolutional neural network model.

[0029] In this embodiment, the lightweight convolutional neural network model adopts the BlazePose model. A rectangular coordinate system is established, and each frame of the video is placed in the rectangular coordinate system, and the coordinates of any point in the image can be obtained.

[0030] S2: Use the human skeleton point data processed by the sliding window mean filtering algorithm to construct a fusion feature vector based on static features and dynamic features.

[0031] Specifically, the skeleton point data is N, and the sl...

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 relates to the field of action recognition, and discloses a human body training action recognition and counting method and system, and the recognition and counting method comprises the following steps: collecting a human body training video, and extracting the skeleton point data of a human body in the video through a lightweight convolutional neural network model; constructing a fusion feature vector based on static features and dynamic features by using the human skeleton point data processed by a sliding window mean filtering algorithm; performing action recognition on a time sequence corresponding to the fusion feature vector by using a bidirectional long-short-term memory neural network; according to characteristics corresponding to different actions during human body training, performing action counting by adopting a difference selection peak value method; the device can run on a mobile phone in real time, and can identify and count the endurance quality instrument-free training actions of the teenagers without depending on other equipment.

Description

technical field [0001] The invention relates to the field of action recognition, in particular to a method and system for recognizing and counting human body training actions. Background technique [0002] The results of the National Student Physical Fitness and Health Survey show that in 2019, 23.8% of the national students aged 6 to 22 years old achieved good and good physical health. Compared with 2014, the good and good rate of high school students only increased by 1.8%. It can be seen that the physical health level of young people in my country is rising slowly, the rate of good and good is low, and the overall situation is not optimistic. Among them, endurance quality, as one of the basic physical qualities of the human body, is a comprehensive reflection of the cardiorespiratory endurance and cardiovascular function level of the human body. Improving endurance quality is very important for physical enhancement. In order to improve the endurance quality of young peop...

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): G06V40/20G06N3/04G06N3/08G06V10/764G06V10/80G06V20/40
CPCG06V40/23G06V20/41G06V20/46G06V10/806G06V10/765G06N3/08G06N3/045
Inventor 孙少明王君洪彭伟孙根基王俊田新壮
Owner 长三角信息智能创新研究院