Supercharge Your Innovation With Domain-Expert AI Agents!

A vehicle re-identification method based on dual-relational attention mechanism

An attention and dual relation technology, applied in the field of vehicle information re-identification, to achieve the effect of improving efficiency and accuracy, optimizing inter-class similarity, and good performance

Active Publication Date: 2022-03-29
SHANDONG JIAOTONG UNIV
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The attention mechanism can capture the key information for vehicle recognition and eliminate the interfering information, but most of the existing attention mechanisms use the information of the feature nodes themselves to measure their importance, or only consider the feature nodes in a certain The importance of dimensions (channels, spaces), with certain limitations

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 vehicle re-identification method based on dual-relational attention mechanism
  • A vehicle re-identification method based on dual-relational attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0019] Such as figure 1 As shown, a kind of vehicle re-identification method based on the double-relationship attention mechanism involved in this embodiment includes the following steps:

[0020] Step 1. Use ResNet-50 as the backbone network, and split ResNet-50 into three branches (global branch G1, global branch G2, and global branch G3). The image is input into the backbone network, and the vehicle's image is extracted through the three branches. different characteristics.

[0021] The specific method of step 1 is as follows: figure 1 As shown, the ResNet-50 convolutional neural network architecture is used as the backbone network, and the second half of ResNet-50 is split into three branches. The second half of ResNet-50 refers to the second block of the 4th layer of ResNet-50 And the following part, the first half of ResNet-50 refers to the first block and the previous part of the 4th layer of ResNet-50. The image is scaled to a uniform size of 256×256, input into the...

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 fields of image processing, computer vision and pattern recognition, and in particular relates to a vehicle re-identification method based on a dual-relationship attention mechanism, which uses the dual-relationship attention mechanism to obtain a large amount of saliency information for vehicle re-identification . The described method mainly consists of two parts: dual-relational attention mechanism and non-similarity constraints. The dual-relational attention mechanism is to construct the three-dimensional attention of the points on the feature map through the "relationship", that is, to construct the importance of a certain point in the spatial dimension and the channel dimension at the same time, to comprehensively measure the importance of the point; non-similarity Sexual constraints make different branches of the network focus on different discriminative information to capture rich and distinguishable saliency information, thereby improving the efficiency and accuracy of the vehicle re-identification task.

Description

technical field [0001] The invention belongs to the technical fields of image processing, computer vision and pattern recognition, and in particular relates to a method for vehicle information re-identification. Background technique [0002] As an important part of intelligent transportation and urban surveillance systems, vehicle re-identification has received more and more attention in recent years. The purpose of the vehicle re-identification task is to obtain images or video sequences from multiple cameras in different positions, and retrieve a specific vehicle from them, that is, given a picture of a query vehicle, it can be found from the gallery by different cameras from different directions in different scenes. Captured vehicles are matched. [0003] Vehicle re-identification has gained rapid attention in the past few years due to its wide applications in urban surveillance and intelligent transportation. Among them, the latest progress in vehicle re-identification...

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): G06V20/40G06V20/54G06V10/74G06V10/82G06K9/62G06N3/04
CPCG06N3/045G06F18/22
Inventor 王成郑艳丽姜刚武田鑫庞希愚李曦孙珂郑美凤周厚仁
Owner SHANDONG JIAOTONG UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com