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

Human body posture recognition method and device based on grouping convolution, equipment and medium

A technology of human posture and recognition method, which is applied in the field of computer vision, can solve problems such as low recognition efficiency, long convergence time, and high complexity of the human body, so as to improve the accuracy rate, avoid inaccurate posture recognition, and improve the overall effect

Pending Publication Date: 2020-10-09
北京软通智慧科技有限公司
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although effective in recognition accuracy, features extracted from bone joint points and depth image information have certain limitations due to the high complexity of the human body
And most of the depth images need to be preprocessed, resulting in difficulty in pose feature extraction, low recognition efficiency, and long convergence time

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 posture recognition method and device based on grouping convolution, equipment and medium
  • Human body posture recognition method and device based on grouping convolution, equipment and medium
  • Human body posture recognition method and device based on grouping convolution, equipment and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] figure 1 It is a flow chart of the human body posture recognition method based on group convolution in Embodiment 1 of the present invention. This embodiment is applicable to the situation of performing human body posture recognition on human body images. The method can be performed by a device for human gesture recognition based on group convolution, which can be implemented in the form of software and / or hardware, and can be configured in a device, for example, the device can be a background server and other devices with communication and computing capabilities . like figure 1 As shown, the method specifically includes:

[0027] Step 101, acquire an image to be recognized.

[0028] The image to be recognized refers to human moving image data including the whole or part of the human body, for example, including the overall outline of the human body and gesture limbs. The images to be recognized can be monitoring images of road pedestrians, human body motion images,...

Embodiment 2

[0043] Figure 2A It is a flow chart of the human body posture recognition model training process in the second embodiment of the present invention. The second embodiment of the present invention describes the training process of the human body posture recognition model in the first embodiment of the present invention in detail. The second embodiment can be set in the embodiment before the identification process. like Figure 2A As shown, the training process includes:

[0044] Step 201. Acquire sample set images.

[0045] The image of the sample set is determined from the pre-acquired sample set. Exemplarily, the image data is collected using the data collection function in the intelligent image analysis platform to obtain a sample set representing human activity image data, and a pose label is added to the sample set .

[0046] And because the convolutional neural network is a feed-forward neural network, since the network avoids the complex preprocessing of the image, i...

Embodiment 3

[0083] image 3 It is a schematic structural diagram of a device for recognizing human body posture based on group convolution in Embodiment 3 of the present invention. This embodiment is applicable to the situation of performing human body posture recognition on human body images. like image 3 As shown, the device includes:

[0084] An image acquisition module 310, configured to acquire an image to be identified;

[0085] The output result determination module 320 is used to input the image to be recognized into the human body posture recognition model to obtain the output result of the human body posture recognition model; wherein, the human body posture recognition model includes at least two groups of convolution units for obtaining the image to be recognized At least two characteristic data of ;

[0086] The pose recognition module 330 is configured to determine the pose of the human body in the image to be recognized according to the output result.

[0087] The embo...

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 embodiment of the invention discloses a human body posture recognition method and device based on grouping convolution, equipment and a medium. The human body posture recognition method based on grouping convolution comprises the steps of acquiring a to-be-recognized image; inputting the to-be-recognized image into a human body posture recognition model to obtain an output result of the humanbody posture recognition model, wherein the human body posture recognition model comprises at least two groups of convolution units and is used for obtaining at least two pieces of feature data of theto-be-recognized image; and determining a human body posture in the to-be-recognized image according to the output result. According to the embodiment of the invention, the globality of human body posture feature extraction is improved through the at least two feature data of the to-be-recognized image, the inaccurate posture recognition caused by the impact from local features is avoided, and the posture recognition accuracy and the generalization of the human body posture recognition model are improved.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of computer vision, and in particular to a method, device, device and medium for human gesture recognition based on group convolution. Background technique [0002] In recent years, with the development of information technology and the popularity of artificial intelligence technology, human gesture recognition technology has begun to be widely used. Related researchers use the collected human pose datasets to explore effective features and classify them. Traditional pose recognition methods mainly have two steps: (1) Extract complex artificial features from the original input image; (2) Train a classifier from the acquired features. [0003] In the traditional pose recognition process, complex artificial features need to be extracted from the original input image. Although effective in recognition accuracy, features extracted from bone joint points and depth image information hav...

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
IPC IPC(8): G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06V40/103G06V10/454G06N3/045G06F18/24G06F18/214
Inventor 袁振杰郝瑞雒冬梅李慧强宋亚莲
Owner 北京软通智慧科技有限公司
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