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

Gesture recognition method and system based on skeleton

A gesture recognition and skeleton technology, applied in the field of computer vision, to achieve recognition efficiency and recognition accuracy improvement, good smoothing effect, and space saving effects

Active Publication Date: 2020-06-16
SHANDONG UNIV
View PDF7 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The motion features extracted by manual features describe the relationship between different joints of gestures in different forms, but manual features have limited generalization ability, and various existing deep learning methods are still unable to extract rich gesture motions well. feature

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
  • Gesture recognition method and system based on skeleton
  • Gesture recognition method and system based on skeleton
  • Gesture recognition method and system based on skeleton

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] In one or more implementations, a recognition method based on skeleton dynamic gestures is disclosed, such as figure 1 shown, including the following steps:

[0043] Step 1: Obtain the skeleton data of the gesture;

[0044] Step 1 specifically includes:

[0045] Step 1.1: Read the TXT file that stores the skeleton data, and each line in the file represents the skeleton data of one frame. Each frame of skeleton data is represented by 22 bone nodes, which are stored in the format of .

[0046] Step 1.2: Transform skeleton data into a format suitable for network input. The input data of the network is in the form of a matrix, each row represents the skeleton data of each frame, we need to store it in the following format:

[0047] 1 ,y 1 ,z 1 ),(x 2 ,y 2 ,z 2 ),…,(x 21 ,y 21 ,z 21 ),(x 22 ,y 22 ,z 22 )>

[0048] Specifically, this embodiment can realize real-time gesture skeleton data acquisition by using Leap motion on the one hand, and realize gesture rec...

Embodiment 2

[0151] In one or more embodiments, a terminal device is disclosed, including a server, the server includes a memory, a processor, and a computer program stored on the memory and operable on the processor, and the processor executes the The program realizes the skeleton-based dynamic gesture recognition method in the first embodiment. For the sake of brevity, details are not repeated here.

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 gesture recognition method and system based on a skeleton. The gesture recognition method comprises the steps of conducting data enhancement on an obtained to-be-recognized original gesture skeleton sequence; respectively extracting motion features between skeleton nodes in each frame and spatial motion features of different scales, and obtaining a first dynamic gesture prediction label by utilizing a spatial perception network; respectively extracting motion characteristics between adjacent interframe skeleton nodes and time motion characteristics of different scales, and obtaining a second dynamic gesture prediction label by utilizing a short-term time perception network; respectively extracting motion characteristics between non-adjacent interframe skeleton nodes and time motion characteristics of different scales, and obtaining a third dynamic gesture prediction label by using a long-term time perception network; and according to the obtained dynamic gesture prediction label, outputting a final gesture prediction label by utilizing a space-time multi-scale chain network model. According to the invention, improvement of the overall identification efficiency and the identification precision can be realized by optimizing the individual branches in a targeted manner.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a skeleton-based dynamic gesture recognition method and system. Background technique [0002] This section merely sets forth background technical information related to the present invention, which does not necessarily constitute prior art. [0003] With the rapid development of virtual reality technology and the continuous improvement of people's requirements for human-computer interaction performance, the traditional human-computer interaction mode can no longer meet people's needs, and people urgently need more natural interaction methods that are more in line with people's living habits. For example: voice, gesture, etc. At present, gestures have been widely used in the field of human-computer interaction due to their convenience and intuition. They are widely used in various fields such as interactive games, robot control, and assisted communication for deaf-mute pe...

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): G06K9/00G06K9/40G06K9/46G06K9/62G06N3/04G06N3/08G06T5/00
CPCG06N3/08G06T2207/10004G06T2207/20081G06T2207/20084G06V40/28G06V10/30G06V10/44G06N3/045G06F18/2415G06F18/241G06T5/70
Inventor 周元峰李扬科
Owner SHANDONG UNIV
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