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

A hand tracking method and advertising machine

A technology for advertising machines and hands, applied in the field of hand tracking methods and advertising machines, can solve the problem of low universality of hand tracking algorithms in advertising machines, achieve the effect of reducing background dependence and improving stability

Active Publication Date: 2022-05-17
SHENZHEN INTELLIFUSION TECHNOLOGIES CO LTD
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Embodiments of the present invention provide a hand tracking method and an advertising machine to improve the stability of the hand tracking algorithm, thereby solving the problem of low universality of the hand tracking algorithm in the advertising machine

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 hand tracking method and advertising machine
  • A hand tracking method and advertising machine
  • A hand tracking method and advertising machine

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0060] In an optional implementation manner, the training of the single-target multi-box network model includes:

[0061] Obtain a hand image data set, the hand image data set includes hand images under various environments and / or light sources, and the hand images include matting annotations of hands;

[0062] The single-target multi-box network model is trained by using hand images including hand matting labels, and the training process of each hand image includes learning the multi-scale relationship corresponding to the hand in the hand image.

[0063] Among them, the above-mentioned hand image data set can be collected and processed by the user himself. For example, the user obtains the scene where the advertising machine is installed, collects the image in the installed scene, and makes the keying mark of the hand; To call, for example, you can download the hand image data in the EgoHands dataset as the hand image dataset for training. The hand image in each of the abov...

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

An embodiment of the present invention provides a hand tracking method and an advertising machine. The method includes: inputting the obtained input image into a pre-trained single-target multi-box network model for prediction, and calculating the output based on the predicted hand features. The target hand feature position of the image, the input image includes the hand, and the single-target multi-box network model uses the hand as the target to predict during the prediction process; the hand feature position is mapped to the corresponding position of the advertising machine interface and update the position of the cursor tool in the advertising machine interface based on the mapping position, and there is a preset mapping relationship between the output image and the advertising machine interface. Due to the single-target multi-box prediction of hand features, only the hand features need to be extracted as the target for prediction, and there is no need to extract the image background texture features for hand feature prediction, thereby reducing the dependence on the background and improving hand tracking. The stability is conducive to the realization of the landing of the advertising machine.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a hand tracking method and an advertising machine. Background technique [0002] At present, with the vigorous development of deep learning technology in the field of computer vision, it has brought new vitality to commercial applications for related electronic products. Body movements and facial movements are recognized. At present, in the process of human-computer interaction, the position tracking of the recognition part is to distinguish the hand from the background through the texture characteristics of the background of the picture, which requires high background conditions. Once the background or light source changes, the accuracy of the tracking calculation result will be affected. , making human-computer interaction difficult. The solution is to redesign the algorithm after the background or light source changes, which is not conducive to the realization of human-...

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): G06F3/01G06V40/20
Inventor 冯展鹏黄轩王孝宇
Owner SHENZHEN INTELLIFUSION TECHNOLOGIES CO LTD
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