3d convolutional neural network based gesture recognition method

A technology of convolutional neural network and gesture recognition, applied in the field of gesture recognition based on 3D convolutional neural network, to achieve the effect of improving accuracy, improving rationality and reliability, and improving robustness

Active Publication Date: 2018-06-22
JILIN UNIV
View PDF2 Cites 28 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Now many cars are equipped with navigation systems. Going home is the driver’s default setting. When we go home, we need to set up the navigation to go home through the mobile phone or car navigation. It is easy to cause traffic accidents due to distraction.

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
  • 3d convolutional neural network based gesture recognition method
  • 3d convolutional neural network based gesture recognition method
  • 3d convolutional neural network based gesture recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] Include the following steps:

[0038] (1) Collect four types of gesture video data under different lighting conditions

[0039] A camera is used to shoot grayscale and depth video images with a pixel size of 171×224. The data is collected under different lighting conditions, and the gestures of the driver’s right hand are all collected. Among them: the first type of gesture is the left and right sliding gesture, and the second type of gesture is the up and down flip gesture , the third type of gesture is the left and right click gesture, and the fourth type of gesture is the rotation gesture;

[0040] (2) Gesture video data preprocessing

[0041] 1) The time length of the original gesture video is different. In order to unify the time length of each gesture video, the video is regularized, and the nearest neighbor interpolation is used to complete the resampling process by discarding or repeating frames. The result is that the time length of each gesture sequence is i...

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 belongs to gesture recognition methods and relates to a 3d convolutional neural network based gesture recognition method. A series of preprocessing is carried out aiming at common videosand depth videos shot by a depth camera. On the basis of application of a common processing method for preprocessing, a convolutional neural sub-network and a deconvolution neural sub-network combined denoising method is adopted aiming at the problem of noise points of video images, and a 3d convolutional neural network is adopted for processing aimed at a temporal-spatial relationship in the videos. By the 3d convolutional neural network based gesture recognition method, the gesture classification rate is substantially increased, and recognition basis reliability and result reasonability areimproved.

Description

technical field [0001] The invention belongs to a gesture recognition method. Background technique [0002] In a rapidly developing society, communication between people does not rely solely on verbal communication. When we encounter problems that cannot be communicated through language, we will instinctively choose body language. Gestures are one of the important ways of communication between people in our daily life. One, we have been able to express our hospital more accurately. As a research hotspot in the field of computer applications and artificial intelligence, gesture recognition technology is becoming more and more perfect. Mature gesture recognition technology can be applied to the fields of robot control, speech recognition, unmanned driving and motion detection. However, due to the diversity and uncertainty of gestures in time and space, gestures become a complex deformable body, so the current theory of human-computer interaction is not mature enough, and the ...

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 Applications(China)
IPC IPC(8): G06K9/00G06N3/04
CPCG06V40/113G06V40/28G06N3/045
Inventor 许骞艺秦贵和姜慧明张钟翰晏婕刘毅袁帅秦俊
Owner JILIN 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