A Human Behavior Recognition Method Based on RGB-D Video

A recognition method and behavior technology, applied in the field of behavior recognition of computer vision, can solve problems such as stability and low recognition accuracy, and achieve the effect of avoiding dimensional disaster, eliminating weakening effects, and improving accuracy.

Active Publication Date: 2017-07-28
NANJING UNIV OF POSTS & TELECOMM
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] To sum up, the existing human behavior recognition technology based on RGB-D data generally has the problem of feature stability and low recognition accuracy.

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
  • A Human Behavior Recognition Method Based on RGB-D Video
  • A Human Behavior Recognition Method Based on RGB-D Video
  • A Human Behavior Recognition Method Based on RGB-D Video

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] The technical solution of the present invention will be described in detail below in conjunction with the drawings:

[0043] The idea of ​​the present invention is to extract dense MovingPose features, SHOPC features and HOG3D features from RGB-D video acquired by RGB-D cameras according to different motion information, appearance geometric information and texture information according to different human behavior categories, using the MCMKL method to perform The fusion processing of the three characteristics obtains the distinctive complementary information of each characteristic. Finally, the Exemplars-SVM classifier is used to judge the human behavior category.

[0044] The human body behavior recognition method based on RGB-D video of the present invention includes a training phase and a testing phase. The overall process is as follows figure 1 Shown.

[0045] The training phase includes the following steps:

[0046] Step A: Obtain RGB-D video samples of each human behavio...

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 behavior recognition method based on RGB-D video, and belongs to the technical field of computer vision behavior recognition. According to the fact that different types of human behaviors in RGB-D videos have different motion information, geometric information and texture information, the method extracts dense MovingPose features, SHOPC features and HOG3D features from RGB-D videos obtained by RGB-D cameras. The edge-limited multi-core learning method performs feature fusion on the three features, and finally uses the Exemplars-SVM linear classifier to judge the human behavior. Compared with the existing technology, the three extracted features used in the present invention have illumination invariance, scale invariance and viewing angle invariance after fusion, and have significant impact on the appearance difference and behavior process difference between action performers. Robustness can improve the recognition accuracy of human behavior to a certain extent.

Description

Technical field [0001] The invention relates to a human body behavior recognition method, in particular to a human body behavior recognition method based on RGB-D video, and belongs to the technical field of computer vision behavior recognition. Background technique [0002] Human behavior recognition can be applied in many aspects, such as intelligent monitoring, human-computer interaction and sports video processing. Based on the input video data, human behavior recognition methods can be divided into three categories, including: methods based on RGB video data, methods based on depth video data, and methods based on the combination of two kinds of data. [0003] Since depth sensors can obtain richer appearance and structure information, depth cameras have received more and more attention in the field of computer vision. In recent years, more and more companies have developed RGB-D cameras, which are characterized by the ability to provide RGB images and depth images in real tim...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06T7/251G06T2207/10016G06T2207/30196G06V40/103G06F18/23213G06F18/2411
Inventor 陈克虎刘天亮
Owner NANJING UNIV OF POSTS & TELECOMM
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