A pedestrian re-identification method based on a hole convolution and attention learning mechanism

A pedestrian re-identification and learning mechanism technology, which is applied in the field of pedestrian re-identification based on hole convolution and attention learning mechanism, can solve the problems of poor practicability and achieve good practicability

Active Publication Date: 2019-05-21
NORTHWESTERN POLYTECHNICAL UNIV
View PDF6 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] In order to overcome the shortcomings of poor practicability of existing pedestrian re-identification methods, the

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 pedestrian re-identification method based on a hole convolution and attention learning mechanism
  • A pedestrian re-identification method based on a hole convolution and attention learning mechanism
  • A pedestrian re-identification method based on a hole convolution and attention learning mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] refer to Figure 1-4 . The specific steps of the pedestrian re-identification method based on the hole convolution and attention learning mechanism of the present invention are as follows:

[0038] Step 1. Design a bottleneck module based on atrous convolution, and connect multiple bottleneck modules in series to form the final backbone network;

[0039] refer to figure 2 The atrous convolution module in will image 3 The 3x3 convolution in the bottleneck module in is replaced by atrous convolution. Hole convolution consists of three convolutional layers, using convolution kernels of 1x1, 3x3, and 1x1 sizes, respectively. The first 1x1 convolution operation reduces the number of channels of the input feature to a quarter, greatly reducing the number of parameters and improving the speed of the model. The second 3x3 convolution keeps the number of channels constant and learns the local structure information of the image. The third 1x1 convolution restores the numb...

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 pedestrian re-identification method based on hole convolution and an attention learning mechanism, which is used for solving the technical problem that the existing pedestrian re-identification method is poor in practicability. The technical scheme includes: firstly, desigining bottleneck modules based on hole convolution, and connecting a plurality of bottleneck modulesin series to form a trunk network; Pre-training the trunk network to obtain a pre-training model; extracting Attention feature maps at different levels of the backbone network, limiting the consistency of the attention feature maps, and learning attention features at different levels autonomously; Training the network by adopting a cross entropy loss function, a triple loss function and an attention feature map constraint loss function; And directly extracting a final feature by utilizing the trunk network, searching a pedestrian image with the minimum distance from the feature of the to-be-searched pedestrian in a pedestrian retrieval library, endowing the identity to the to-be-searched pedestrian, and finishing a re-identification process. According to the method, the convolutional neural network and the attention learning mechanism are combined, pedestrian re-identification can be accurately carried out, and the practicability is good.

Description

technical field [0001] The invention relates to a pedestrian re-identification method, in particular to a pedestrian re-identification method based on hole convolution and attention learning mechanism. Background technique [0002] Pedestrian re-identification refers to the technology of identifying the identity of pedestrians in different camera scenarios, which is a very important part of video surveillance analysis technology. However, due to the complexity of surveillance video, the impact of drastic changes in lighting, weather, viewing angle changes, pedestrian posture, occlusion and other factors, as well as the poor resolution of imaging equipment, it is difficult to identify the same pedestrian under different cameras. With deep learning, major breakthroughs have been made in many computer vision fields such as image classification and object recognition. [0003] Applying deep learning to pedestrian re-identification can deal with the above problems well. General...

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): G06K9/00
CPCY02T10/40
Inventor 袁媛王琦蒋旻悦
Owner NORTHWESTERN POLYTECHNICAL 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