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

Gait recognition system and method based on LSTM network

A gait recognition and network technology, applied in the field of computer vision and deep learning, can solve problems such as ignoring correlation, destroying video integrity, losing timing information, etc., achieving far-reaching research value, maintaining video integrity, and high accuracy Effect

Pending Publication Date: 2019-08-06
XIAN UNIV OF SCI & TECH
View PDF8 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] To sum up, the existing problems in the existing technology are: the existing gait recognition algorithm simply defines it as the matching problem of gait features between pictures, that is, the video is divided into frames, and then the target picture and the video frame are compared. gait recognition
Ignoring the correlation between adjacent frames in the original video, splitting the video into frames also destroys the integrity of the video and makes it lose timing information

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
  • Gait recognition system and method based on LSTM network
  • Gait recognition system and method based on LSTM network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0014] In order to further understand the content, features and effects of the present invention, the following examples are given, and detailed descriptions are given below with reference to the accompanying drawings.

[0015] The structure of the present invention will be described in detail below in conjunction with the accompanying drawings.

[0016] Such as figure 1 As shown, the gait recognition system based on LSTM network that the embodiment of the present invention provides is provided with:

[0017] A CNN network unit 1 for feature extraction of a picture sequence of a reference video;

[0018] Connected with CNN network unit 1, LSTM neural network unit 2 for simulating timing information;

[0019] Connected with LSTM neural network unit 2, convolutional layer unit 3 for local perception;

[0020] Connected with the convolutional layer unit 3, it is used for the similarity measurement sub-network unit 4 for similarity matching.

[0021] From picture to video gait...

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 the technical field of computer vision and deep learning, and discloses a gait recognition system and method based on an LSTM network. The system is equipped with a CNN + LSTMnetwork unit used for the feature extraction of a reference video, wherein the CNN network unit is used for the feature extraction of the reference video, and an LSTM neural network unit is further used for simulating the time sequence information of the video; and a CNN unit used for the feature extraction of the gait image sequence of a person to be identified, wherein the CNN is used for directly extracting the features of a picture of the person to be identified. According to the gait recognition idea from the image to the video, the time sequence information in the video is considered, and the correlation between the adjacent frames in an LSTM analog video is used, so that the video serves as a whole to analyze the gait features of the video. The model directly learns the spatial features and the temporal features of the video sequence in an end-to-end manner and simultaneously learns and optimizes the feature representation and the similarity measurement.

Description

technical field [0001] The invention belongs to the technical field of computer vision and deep learning, and in particular relates to a gait recognition system and method based on an LSTM network. Background technique [0002] At present, most of the existing gait recognition technologies are based on segmenting human motion videos, then extracting gait features from the video sequence, and then matching the gait features between the reference set and the samples to be identified to complete the identification. Its essence is based on the picture-to-picture matching problem. This type of method does not consider the timing information between adjacent frames in the video, and ignores the correlation between adjacent frames. Based on video sequences that are often human motions available in practical application scenarios. [0003] To sum up, the existing problems in the existing technology are: the existing gait recognition algorithm simply defines it as the matching probl...

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): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06V40/103G06N3/044G06N3/045
Inventor 李洪安
Owner XIAN UNIV OF SCI & TECH
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