Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Multi-mode fused unsupervised pedestrian re-identification rearrangement method

A pedestrian re-identification and multi-modal technology, applied in character and pattern recognition, neural learning methods, biological neural network models, etc., can solve problems such as increasing the workload of manual research and judgment, and reducing the degree of automation of video analysis

Active Publication Date: 2021-02-19
SOUTH CHINA UNIV OF TECH
View PDF5 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This processing method not only increases the workload of manual research and judgment, and reduces the automation of video analysis, but also, due to differences in viewing angles and lighting, the appearance characteristics of pedestrians will change greatly, which may cause the provided results to be ranked higher than others. not necessarily more believable

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
  • Multi-mode fused unsupervised pedestrian re-identification rearrangement method
  • Multi-mode fused unsupervised pedestrian re-identification rearrangement method
  • Multi-mode fused unsupervised pedestrian re-identification rearrangement method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0059] In fact, if the pedestrian image recognition results of each monitoring device on the travel path are viewed as a whole, there should be a strong spatio-temporal dependence between them. For example, it is impossible for the same pedestrian to appear in different monitoring devices that are physically far apart at the same time, and the time difference between pedestrians appearing in different monitoring devices must also have a reasonable relationship with the distance between monitoring devices and the pedestrian speed in common sense , the time for pedestrians to appear on the rear monitoring device on the travel path should not be earlier than that of the front monitoring device. However, the spatio-temporal dependencies related to pedestrian images need to know in advance which pedestrian images have been transferred across cameras, and the spatio-temporal dependencies related to pedestrian images constructed under unsupervised conditions will bring more noise.

...

Embodiment 2

[0109] This embodiment also provides a computer storage medium, the computer storage medium stores computer-executable instructions, and the computer-executable instructions can execute the fusion multimodal unsupervised pedestrian re-identification and rearrangement method in the first embodiment above . Wherein, the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a flash memory (Flash Memory), a hard disk (Hard Disk Drive, abbreviation: HDD) or solid-state hard drive (Solid-State Drive, SSD) etc.; The storage medium may also include a combination of the above-mentioned types of memory.

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 multi-modal fused unsupervised pedestrian re-identification rearrangement method. The method comprises the following steps of collecting multi-modal information of pedestrians in a walking process; extracting pedestrian features by using a convolutional neural network model, and calculating visual similarity; constructing image space-time distribution by utilizing the image space-time information; constructing WiFi space-time distribution by utilizing the WiFi information; and the visual similarity, fusing the image space-time distribution and the WiFi space-time distribution, and arranging a pedestrian re-identification sorting result. Multi-modal information is integrated for secondary rearrangement, the method is an effective measure for reducing the search space, and the defect that a traditional pedestrian re-identification method based on visual features is sensitive to the monitoring environment is effectively overcome.

Description

technical field [0001] The invention belongs to the field of multi-modal intelligent security protection, and in particular relates to an unsupervised pedestrian re-identification and rearrangement method fused with multiple modes. Background technique [0002] At present, pedestrian re-identification, also called pedestrian re-identification, aims to quickly and effectively retrieve target persons from massive surveillance videos, and can track target persons, identify people, and locate missing persons. It plays an important role in safe cities. role. Pedestrian re-identification has attracted extensive research due to its huge application value and the existence of challenging problems such as angle, illumination, occlusion and face blur. [0003] The current mainstream person re-identification method mainly uses labeled data sets for model training, but in reality, labeled data consumes a lot of manpower and financial resources and is difficult to obtain. However, 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
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/53G06N3/045G06F18/22G06F18/2415Y02T10/40
Inventor 吕建明林少川梁天保胡超杰莫晚成
Owner SOUTH CHINA UNIV OF TECH
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
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