A multi-view gait recognition method based on an adaptive 3D human motion statistical model

A statistical model and human motion technology, applied in the field of computer vision and pattern recognition, can solve the problems of incomplete multi-view training set and low recognition accuracy, and achieve the effect of solving the problem of inability to deal with occlusion

Inactive Publication Date: 2020-01-21
武汉盈力科技股份有限公司
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problems that the multi-view training set of the existing gait recognition technology is difficult to complete and the recognition accuracy is not high, the present invention proposes a multi-view gait recognition method based on an adaptive three-dimensional human motion statistical model

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 multi-view gait recognition method based on an adaptive 3D human motion statistical model
  • A multi-view gait recognition method based on an adaptive 3D human motion statistical model
  • A multi-view gait recognition method based on an adaptive 3D human motion statistical model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0022] figure 1 The algorithm flow chart of the present invention in the training stage is given.

[0023] The training uses the video of people walking simultaneously captured by the multi-camera system. Extract the person from the image of each frame, and then extract the feature points only for the person area, and perform matching to generate a set of matching points. Each group of matching points is an image point of the same object point on multiple images. Then, based on the principle of the least square error, the three-dimensional coordinates of the object point are solved through collinear equations. In this way, the coordinates of all object space points in the matching point set can be calculated, and the point cloud of each frame of the human body can be generated. Assuming that all cameras shoot at the same speed, and a complete gait c...

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 multi-view gait recognition method based on an adaptive three-dimensional human motion statistical model. The images of the training set of the present invention come from multiple camera systems, and the point cloud of the object is generated through the three-dimensional reconstruction technology of multi-eye vision, and then a three-dimensional statistical human body model is established. The 3D human body model is transformed through a virtual camera projection to obtain a synthetic human body contour binary image under any viewing angle, which is used to further extract various gait features. Based on the 3D human body model, the bone model is established, the degree of freedom of each joint is given a reasonable range, and a statistical 3D human motion statistical model is established. Through parameter adjustment, it can adapt to various walking occasions. In the training phase, a gait feature database is established based on the above method. In the recognition stage, the same gait features are extracted from the video and compared with the features in the database, and the best recognition object is found through the nearest specimen classifier combined with the highest score strategy.

Description

technical field [0001] The invention relates to computer vision and pattern recognition, in particular to a gait recognition method based on an adaptive three-dimensional human motion statistical model. Background technique [0002] Gait recognition technology is a kind of biometric identification technology, that is, to identify individuals through each person's unique walking style. Compared with the first generation of biometric recognition technology, such as fingerprint recognition, face recognition, iris recognition, etc., gait recognition technology has the advantages of no need for physical contact, low requirements for image resolution, and long-distance recognition. So far, gait characteristics may be the only biological characteristics that can be recognized remotely. Therefore, gait recognition technology has broad commercial application prospects in security monitoring and other fields. [0003] With the development of science and technology and the improvement...

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): G06K9/00G06T7/55G06T7/246
CPCG06T2207/10016G06T2207/30196G06T2207/20081G06V40/25
Inventor 巨辉杨斌曹顺
Owner 武汉盈力科技股份有限公司
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