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

Pedestrian re-identification method based on human body posture invariant features

A technology for pedestrian re-identification and human body posture, applied in character and pattern recognition, image data processing, instruments, etc., can solve the problem of low robustness of pedestrian features, achieve enrichment of local color information, reduce impact, and be affected by lighting conditions small effect

Inactive Publication Date: 2019-07-05
CHANGSHU INSTITUTE OF TECHNOLOGY
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In short, the problems existing in the existing technology are: the effect of pedestrian re-identification is easily affected by changes in lighting conditions and human body posture, and the robustness of pedestrian features is low.

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
  • Pedestrian re-identification method based on human body posture invariant features
  • Pedestrian re-identification method based on human body posture invariant features
  • Pedestrian re-identification method based on human body posture invariant features

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0061] 1. Dataset

[0062] In order to verify the effectiveness of the method of the present invention in improving the accuracy of pedestrian re-identification, according to the above steps, experiments were carried out on three datasets of pedestrian re-identification, namely VIPeR, CUHK03 and Market1501.

[0063] Among them, VIPeR is the earliest and one of the most classic data sets proposed in the field of pedestrian re-identification. It includes 632 pedestrians and has 1264 images. Each pedestrian corresponds to an image under two cameras, and each image is manually labeled. In addition to this, the dataset also provides a rich variety of perspectives, and although the dataset is relatively old, it is still one of the most challenging datasets.

[0064] CUHK03 was collected from 10 cameras at the Chinese University of Hong Kong. The dataset contains 1467 pedestrians and 14096 images. Each pedestrian comes from at least two different cameras, with an average of 4.8 ped...

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 human body posture invariant features, which comprises the following steps of: (10) original image preprocessing: preprocessing an original pedestrian image by utilizing a multi-scale retina enhancement algorithm with color recovery factors to obtain a corrected image with rich local color information; (20) model training: taking the original pedestrian image and the correction image as input, and training a ResNet 50-based pseudo twin neural network model; (30) invariant feature extraction: extracting human body posture invariant features by using the trained pseudo-twin neural network model; and (40) pedestrian re-identification: re-identifying the pedestrian according to the human body posture invariant feature. According to the pedestrian re-identification method based on the human body posture invariant features, the identification effect is less influenced by illumination conditions and human body posture changes.

Description

technical field [0001] The invention belongs to the technical field of intelligent surveillance video recognition, in particular to a pedestrian re-identification method based on the invariant feature of human body posture which is less affected by illumination conditions and changes in human body posture. Background technique [0002] Pedestrian re-identification is a cross-camera pedestrian recognition technology. Generally speaking, pedestrian re-identification includes two parts: target detection and target recognition. Among them, target detection refers to the process of locating and extracting pedestrians from the original video. Target recognition refers to the process of retrieving the corresponding results from the candidate set for the specified target in a cross-view environment. Finally, person re-identification returns a ranked list of candidate images sorted by their similarity (distance) to the target image. [0003] The emergence of pedestrian re-identifi...

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): G06K9/00G06K9/46G06K9/62G06T3/00G06T5/00
CPCG06T2207/30196G06T2207/20081G06T2207/20084G06T2207/20021G06V40/103G06V10/464G06F18/24G06T3/02G06T5/94
Inventor 龚声蓉包宗铭
Owner CHANGSHU INSTITUTE OF TECHNOLOGY
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