Real-time distributed identity perception pedestrian attribute identification method in mobile edge cloud

An attribute recognition and pedestrian re-identification technology, applied in the field of identity-aware pedestrian attribute recognition, can solve problems such as communication overhead and computing power limitation, and achieve the effect of improving accuracy, reducing model capacity and facilitating distribution.

Active Publication Date: 2020-08-18
DALIAN UNIV OF TECH
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing CV model reasoning framework cannot be directly deployed in the MEC network, and there are problems such as computing power limitations and communication overhead. Therefore, it is urgent to integrate the CV reasoning model and task assignment in the MEC network at a finer granularity.

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
  • Real-time distributed identity perception pedestrian attribute identification method in mobile edge cloud
  • Real-time distributed identity perception pedestrian attribute identification method in mobile edge cloud
  • Real-time distributed identity perception pedestrian attribute identification method in mobile edge cloud

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050] In order to express the purpose, technical solutions and advantages of the present invention more clearly, the present invention will be further described in detail through the following examples and accompanying drawings. The specific embodiments here are only used to illustrate the present invention, but not to limit the present invention.

[0051] A real-time distributed identity-aware pedestrian attribute recognition method and system in a mobile edge cloud. The system used includes a neural network training subsystem, a distributed reasoning subsystem, and a context-based multi-armed bandit machine-based module allocation subsystem.

[0052] refer to figure 1 , the detailed description of the neural network training subsystem is as follows:

[0053] ResNet-50 pre-trained on ImageNet is used as the backbone network to extract feature representation from the input image;

[0054] The existing joint training methods of pedestrian attribute recognition and pedestrian...

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 provides a real-time distributed identity perception pedestrian attribute identification method in mobile edge cloud. According tothe invention, a system for analyzing pedestrian attributes bound with identities across cameras is provided for the first time, and a complete algorithm implementation and deployment scheme is provided. In addition, some local algorithms have improvements on the prior art. Firstly, an improved pedestrian attribute recognition and pedestrian re-recognition combined training method improves the recognition precision while reducing the redundant information of the neural network; and 2, a distributed pedestrian image library, wherein the process of feature distance calculation in pedestrian re-identification is accelerated by adopting parallel calculation. In addition, the module allocation algorithm based on the context multi-arm tiger machine has universality and can be used for deployingother models in a mobile edge network.

Description

technical field [0001] The invention relates to pedestrian attribute recognition and pedestrian re-identification technology, belongs to the field of computer vision, and relates to a real-time distributed identity-aware pedestrian attribute recognition method in a mobile edge cloud. Background technique [0002] The purpose of Pedestrian Attribute Recognition (PAR) technology is to identify the attributes of target pedestrians, such as gender, age, hairstyle, clothing, etc., from a given image or video sequence. Pedestrian attributes are semantic descriptions that can be used for retrieval, and are increasingly being used in the field of intelligent monitoring. Different from low-level features such as Histogram of Oriented Gradient (HOG), attributes can be regarded as high-level semantic information, which is usually integrated into other tasks in the field of computer vision as auxiliary information. After the development of pedestrian attribute recognition from the trad...

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): G06K9/00G06K9/62G06N3/04G06N3/08G06N5/04G06F9/50
CPCG06N3/08G06N5/04G06N5/043G06F9/5027G06F2209/502G06V40/103G06N3/045G06F18/241
Inventor 徐子川吴将凯夏秋粉
Owner DALIAN 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
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