Multi-person posture estimation method based on deep cascade network and centroid differentiation coding

A cascaded network and attitude estimation technology, applied in reasoning methods, calculations, computer components, etc., can solve problems such as differences, human joint occlusion scales, etc.

Active Publication Date: 2019-11-08
HUAQIAO UNIVERSITY +1
View PDF11 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to provide a multi-person pose estimation method base...

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
  • Multi-person posture estimation method based on deep cascade network and centroid differentiation coding
  • Multi-person posture estimation method based on deep cascade network and centroid differentiation coding
  • Multi-person posture estimation method based on deep cascade network and centroid differentiation coding

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] Such as figure 1 As shown, the present invention is based on the depth cascade network and the multi-person attitude estimation method of centroid differentiation encoding, including:

[0053] Step 1. Establish a deep cascaded network and perform training;

[0054] Step 2. Use the trained deep cascade network to calculate an image to be tested, and obtain all human joint points and corresponding centroid differentiation codes. The centroid differentiation codes are the centroid positions of the human body half to which the joint points belong; based on the centroid differentiation codes, Carry out greedy reasoning on all joint points, and combine the joint points to obtain multiple human upper and lower bodies respectively;

[0055] Step 3. Add space constraints according to the joint information in the upper body and lower body, and then use the bipartite graph matching algorithm to combine the upper body and lower body to finally obtain the complete posture of multip...

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 provides a multi-person posture estimation method based on a deep cascade network and centroid differentiation coding. An estimation route from bottom to top is adopted; the problems ofjoint shielding and scale difference which cannot be solved by an existing algorithm are solved. A centroid differentiation code is designed as a correlation clue of the joint; a deep cascade networkbased on a two-way feature extraction module is established to complete extraction of joint point and centroid differentiation codes; and then a greedy reasoning strategy is proposed to realize robustmatching of joint points to multiple human body halves, finally spatial constraints are added between the halves, and human body splicing is completed by using a graph matching algorithm so as to realize rapid and efficient multi-person posture estimation.

Description

technical field [0001] The invention relates to the field of human pose estimation in computer vision, in particular to a multi-person pose estimation method based on deep cascade network and centroid differentiation coding. Background technique [0002] Human body pose estimation is a key step in the design and manufacture of smart devices to understand human behavior. The purpose is to locate and identify the joint points of all human bodies in the image and connect them to form a human skeleton. Effectively predicting human joint points and obtaining corresponding human poses is of great significance for realizing higher-level computer vision tasks such as advanced human-computer interaction, behavior recognition, and pedestrian re-identification. Although there are many studies on pose estimation technology, the existing multi-person pose estimation technology is far from mature, and it is still facing great challenges to fully realize robust and high-precision multi-per...

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): G06K9/00G06K9/62G06N5/04
CPCG06N5/04G06V40/103G06F18/214
Inventor 骆炎民张智谦林躬耕缑锦
Owner HUAQIAO UNIVERSITY
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