A kind of hand area tracking method and system

A hand and region technology, applied in the field of gesture recognition, can solve problems such as the inability to guarantee the tracking accuracy of the hand region, the large gap in real-time accuracy of the mobile AR assembly system, large storage space, and high hardware configuration in the operating space, etc., to achieve Accurate tracking, reduced impact, and improved tracking speed

Active Publication Date: 2022-07-29
LANZHOU JIAOTONG UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although SLAM technology and deep learning methods can improve the stability of 3D registration, a large number of parameters in its complex model structure require a large storage space, running space and high hardware configuration, making the real-time accuracy of the mobile AR assembly system comparable to the actual Big gap in demand
Therefore, in the mobile AR assembly system based on fingertip interactive input, the use of SLAM technology and deep learning methods cannot guarantee the accuracy of hand area tracking

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 kind of hand area tracking method and system
  • A kind of hand area tracking method and system
  • A kind of hand area tracking method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0064] The tracking and registration method of mobile AR assembly system with natural fingertip interaction can provide support for the application of natural fingertip interaction in AR assembly system through adaptive correlation filtering of hand area tracking, making natural video sequence hand area tracking a The natural fingertips move the basis of the AR assembly system, so it is necessary to accurately track the hand area in real time even in complex environments such as hand deformation, illumination, scale, and rotation, and at the same time solve the target template that exists in the hand area tracking. Drift problem has become a key problem that needs to be solved in the natural fingertip mobile AR assembly system. Existing SLAM technology and deep learning methods cannot guarantee the real-time and accuracy of hand area tracking.

[0065] In this regard, this embodiment 1 provides a hand area tracking method, which utilizes the interactive control theory and the ...

Embodiment 2

[0137] see image 3 , this embodiment also provides a hand area tracking system, including:

[0138] The acquisition module M1 is used to acquire two-dimensional video sequence images in the mobile AR assembly system;

[0139] The current frame image determination module M2 is used to determine the spatial scale and position of the hand region in the two-dimensional video sequence image of the current frame, and obtain images of corresponding positions under different spatial scales;

[0140] The training sample determination module M3 is used to use the images of the corresponding positions under the different spatial scales as the training samples;

[0141] The training module M4 is used to train the KCF tracker using the training sample to obtain the trained KCF tracker;

[0142] The next frame image determination module M5 is used to determine the spatial scale of the hand region in the two-dimensional video sequence image of the next frame, and obtain target tracking im...

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 relates to a hand area tracking method and system. The method first collects two-dimensional video sequence images from a mobile AR assembly system interacting with natural fingertips, and preprocesses the collected images in terms of image size and transformation angle. Using the method of hand skin color mean iterative segmentation and APBS background subtraction to determine the position of the hand area to be registered, and realize the location of the hand area; then apply the correlation filter tracking algorithm to the hand area tracking, and adopt a strategy that considers all previous frames. To solve the weight vector update problem of each frame of image training of the KCF algorithm, realize the adaptive tracking of the hand region by the KCF algorithm, and then realize the real-time, accurate and stable tracking of the hand region in complex environments, and provide accurate real-time detection and recognition of fingertips. support.

Description

technical field [0001] The present invention relates to the technical field of gesture recognition, and in particular, to a hand area tracking method and system. Background technique [0002] The mobile AR assembly system emphasizes the synthesis and interaction of 3D virtual objects and the real world, and the 3D registration technology is the basis, key and difficulty of building a mobile AR system. At present, the main development trend of 3D registration is the SLAM technology of visual tracking and the deep learning method. Although SLAM technology and deep learning methods can improve the stability of 3D registration, the large number of parameters existing in its complex model structure requires huge storage space, running space and high hardware configuration, which makes the real-time accuracy of mobile AR assembly system and practical The demand gap is large. Therefore, in the mobile AR assembly system based on the interactive input of fingertips, the use of SLAM ...

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): G06F3/01G06V10/764G06V20/40G06V10/774G06V40/10G06K9/62
CPCG06F3/017G06F18/214G06F18/24
Inventor 雍玖王阳萍党建武雷晓妹杨景玉王松
Owner LANZHOU JIAOTONG UNIV
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