Behavior recognition method and system based on human skeleton

A technology of human skeleton and recognition method, applied in the field of computer vision, can solve problems such as reduction, and achieve the effect of reducing the amount of calculation, improving the accuracy, and increasing the connection

Pending Publication Date: 2022-04-19
FUDAN UNIV
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The existing spatio-temporal graph convolutional network model effectively utilizes spatial features and temporal features, and classifies two behavior recognition data sets through multi-task training methods to reduce overfitting, but the accuracy caused by inappropriate shooting angles The impact is not dealt with

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
  • Behavior recognition method and system based on human skeleton
  • Behavior recognition method and system based on human skeleton
  • Behavior recognition method and system based on human skeleton

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] A behavior recognition method based on human skeleton, such as figure 1 shown, including the following steps:

[0051] S1. Acquire temporally continuous image sequences and perform preprocessing. The images in the image sequence are human behavior images;

[0052] It can acquire video or multiple images that are continuous in time, and sample them according to the preset sampling interval, such as the time interval between adjacent frames of images is 0.1s, etc., to obtain a sequence of continuous images in time.

[0053] Image preprocessing includes common denoising, delineating the region of interest where the human body is located, and image scaling, cropping, etc. Denoising can reduce the influence of interference factors. The region of interest where the human body is located can be manually delineated, or the corresponding artificial Intelligent detection algorithm for identification, image scaling and cropping to process the image to an appropriate size, and ada...

Embodiment 2

[0076] This application also protects a human skeleton-based behavior recognition system, including:

[0077] The data acquisition module acquires temporally continuous image sequences and performs preprocessing, and the images in the image sequences are human behavior images;

[0078] The skeleton extraction module uses the pose estimation method to obtain the skeleton information in each image, and obtains the skeleton sequence corresponding to the image sequence;

[0079] The feature extraction module sends the skeleton sequence to the first model and the second model respectively. The first model is a temporal convolutional network for extracting temporal features of the skeleton sequence. The second model includes a view adaptive network and a spatial graph convolutional network. , used to extract the spatial features of the skeleton sequence, the output of the view adaptive network is sent to the spatial graph convolutional network;

[0080] The feature fusion module in...

Embodiment 3

[0083] The behavior recognition method based on the perspective optimization of the human skeleton and the fusion of the spatio-temporal graph neural network proposed by the present invention can automatically recognize human behaviors from the video, and can be widely used in security active warnings, smoking behavior recognition, and children's fall detection In computer vision applications such as computer vision, it provides method guidance for computer vision task behavior recognition, which has broad application prospects and potential economic and social values.

[0084] In this embodiment, the specific implementation manners of the present invention in the security system, urban civilization monitoring network and fall detection system are given respectively.

[0085] (1) Apply this application to establish an AI behavior monitoring system for intelligent security:

[0086] The AI ​​behavior monitoring system is an intelligent monitoring system that can identify variou...

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 a behavior recognition method and system based on a human skeleton. The method comprises the following steps: acquiring an image sequence which is continuous in time; using an attitude estimation method to obtain a skeleton sequence; the skeleton sequence is sent to a first model and a second model, the first model is a time convolutional network and is used for extracting time features of the skeleton sequence, and the second model comprises a view adaptive network and a space graph convolutional network and is used for extracting space features of the skeleton sequence; and fusing the time features and the space features, and outputting human behavior actions. Compared with the prior art, the method has the advantages that the image information is converted into the human skeleton sequence information, the spatial features and the time features are extracted respectively, the visual angle self-adaptive adjustment is performed in the spatial feature extraction, and the connection relationship between the joint points is added, so that the recognition accuracy and robustness are better.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a behavior recognition method and system based on a human skeleton, which integrates perspective optimization and a spatio-temporal graph neural network. Background technique [0002] Human action recognition has become an active research area in recent years, and it plays an important role in video understanding. Behavior recognition has become one of the basic problems in the field of computer vision, and the behavior recognition algorithm based on deep learning is the mainstream algorithm of current behavior recognition. Human behavior is an event that occurs in a certain time and space, with spatial and temporal characteristics. The key issue of behavior recognition is how to effectively describe the temporal and spatial characteristics. From different perspectives, some studies process image space and time information separately through two neural networks, and fin...

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): G06V40/20G06V20/52G06V10/30G06V10/40G06V10/80G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/049G06N3/08G06N3/044G06N3/045G06F18/253
Inventor 张立华魏志强石鑫鑫
Owner FUDAN 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