A pedestrian rerecognition method based on Gaussian and depth features

A pedestrian re-identification and deep feature technology, applied in biometrics, character and pattern recognition, instruments, etc., can solve problems such as not being able to handle small data sets well

Inactive Publication Date: 2019-03-19
中汽数据(天津)有限公司
View PDF2 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the convolutional neural network does not pay special attention to these features, but instead extracts the semantic

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 rerecognition method based on Gaussian and depth features
  • A pedestrian rerecognition method based on Gaussian and depth features
  • A pedestrian rerecognition method based on Gaussian and depth features

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0087] see Figure 1 to Figure 2 Shown:

[0088] The pedestrian re-identification method based on Gaussian and depth features provided by the present invention, its method is as follows:

[0089] Step 1. Find the video of the same road section taken by different cameras from the Internet, and extract frames from the video according to a certain period of time, manually draw the pedestrian's position frame, and collect the training set and test set. The steps are as follows:

[0090] (1) Download the common image data set (Market1501) in pedestrian re-identification technology from the Internet to form an image collection where N d is the total number of images in the collection IMG.

[0091] (2), the data set has a total of 1501 pedestrians, N d = 32668 location boxes, these pedestrians share 6 cameras to capture, and each pedestrian uses at least two cameras to label.

[0092] (3) Use the Deformable Part Model (DPM) technology to extract the location frame, thereby gene...

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 recognition method based on Gaussian and depth features. The method includes 1, collecting a training set and a test set; 2, quickly extracting pedestrian features; 3, extracting semantic features of pedestrian feature; 4, training and extracting feature with that model; 5, testing that performance of the JDAG model; step 6, reordering the final result queue. Beneficial effect: Combining the semantic features extracted by convolution neural network with the unique features (color, texture) of pedestrians, the distinguishability of features is improved. Next, the Re-ranking method rearranges the previous similarity scores so that a picture is given and the pedestrians in the picture set with the same ID as the pedestrians in the picture can be found moreaccurately.

Description

technical field [0001] The invention relates to a pedestrian re-identification method, in particular to a pedestrian re-identification method based on Gaussian and depth features. Background technique [0002] Currently, an emerging task in intelligent supervision systems in the field of computer vision—person re-identification. This task is mainly to solve the problem of pedestrian recognition and pedestrian identity similarity matching in complex and changing environments. The key problem to be solved in pedestrian re-identification is to match the pedestrian pictures in the index queue with the pictures in the entire data set to find the same person. Of course, the pictures of pedestrians in the data set are captured by cameras at different positions and then intercepted, that is, to ensure the authenticity of the pedestrian scene. However, precisely because of the different positions and angles of the cameras, some external factors of pedestrians will bring great chall...

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): G06K9/00G06N20/00
CPCG06V40/10
Inventor 朱向雷杜志彬赵帅张鲁武毅男周博林翟洋陈蔯
Owner 中汽数据(天津)有限公司
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