A method for detecting the position arrangement of key points of human skeleton on a multi-person image

A technology of human body position and detection method, applied in the field of computer vision research, which can solve the problems of high interference and inability to combine the deep noise calculation model well

Active Publication Date: 2019-01-22
SHANGHAI MOSHON TECH CO LTD
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a method for detecting the position arrangement of key points of human skeleton on a multi-person image, which is used to solve the problem of large interference caused by R

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 method for detecting the position arrangement of key points of human skeleton on a multi-person image
  • A method for detecting the position arrangement of key points of human skeleton on a multi-person image
  • A method for detecting the position arrangement of key points of human skeleton on a multi-person image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0038] refer to figure 1 , a method for detecting the position arrangement of human skeleton key points on a multi-person image disclosed in this embodiment includes: inputting the original color image of human skeleton key points into the human body position and texture full convolutional neural network training model f; Position and texture The fully convolutional neural network training model f calculates and generates the human body position and texture color feature map F that highlights the position and texture of each person on the original color image of the key points of human bones; the human body position and texture color feature map F is converted to RGB After the grayscale calculation, the human body position and texture grayscale feature map F' is generated; the human body position and texture grayscale feature map F' are respectively input into the seven-stage human skeleton key point mass information full convolutional neural network training model ρ k And sev...

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 method for detecting the position arrangement of key points of human skeleton on a multi-person image, By constructing and training the full convolution depth neural network,the original color image of human skeleton key points is transformed into depth map after being preliminarily processed and then used as input of the full convolution depth neural network, after several stages of cyclic transformation, the 18-layer human skeleton single-key-point quality trust map and 17-layer human skeleton single-key-point link field map of 17 human skeleton key-point link segments are calculated and outputted, in the calculation of multi-stage cyclic transformation, a loop end verification judgment formula is used to verify that node whose loop ends, in addition, the calculation of the multi-stage cyclic transformation is also trained and controlled by calculating the total loss L of the calculation of the multi-stage cyclic transformation. The invention effectively utilizes the characteristic information and combines the global and local information, thereby outputting more abundant characteristic information and improving the positioning effect of the key pointsof the bone.

Description

technical field [0001] The invention relates to the technical field of computer vision research, in particular to a method for detecting the position arrangement of human skeleton key points on a multi-person image. Background technique [0002] One of the important tasks in the field of computer vision research is the detection of key points of human bones. Specifically, it is to enable computers to perceive the positions of key points of human bones, and provide a basis for further action recognition, action anomaly detection and other practical scenarios. [0003] The goal of the human skeleton key point detection task is to take a picture as input, and output the horizontal and vertical coordinates of each bone key point of the human body in the picture. There are often two types of input images, one is a three-dimensional RGB color image, and the other is a two-dimensional depth map. RGB images are often difficult to detect key points of human bones due to interference...

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): G06T7/00G06T7/73
CPCG06T7/0012G06T7/73G06T2207/20081G06T2207/20084G06T2207/30008G06T2207/30196
Inventor 梁峰浦汉来
Owner SHANGHAI MOSHON TECH CO LTD
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