Method for training zero-sample image classification by using total data

A technology of full data and sample images, applied in computer parts, instruments, character and pattern recognition, etc., can solve the problem of missing image category labels, and achieve the effect of solving the missing category labels, alleviating the problem of strong bias prediction, and increasing the loss.

Active Publication Date: 2020-02-21
FUZHOU UNIV
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of this, the object of the present invention is to provide a zero-sample image classificatio

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
  • Method for training zero-sample image classification by using total data
  • Method for training zero-sample image classification by using total data
  • Method for training zero-sample image classification by using total data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0040] Please refer to figure 1 , the present invention provides a zero-sample image classification model utilizing a full amount of data training, including a visual feature network, an attribute semantic conversion network, a visual-attribute semantic connection network and a score network; specifically comprising the following steps:

[0041] Step S1: dividing the full amount of data into source data and target data;

[0042] Step S2: input the source data set and target data set into the visual feature network, map the original image to the visual feature space, and obtain the image visual feature vector;

[0043] Step S3: the low-dimensional attribute semantics of the original image is mapped to high-dimensional by the semantic conversion network, and obtain the semantic feature vector;

[0044] Step S4: according to the image visual feature vect...

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 relates to a method for training zero-sample image classification by using total data. The method comprises the following steps: S1, dividing the total data into source data and target data; S2, inputting the source data set and the target data set into a visual feature network, and mapping the original image to a visual feature space to obtain an image visual feature vector; S3, mapping the low-dimensional attribute semantics of the original image to a high dimension through a semantic conversion network to obtain a semantic feature vector; S4, completing fusion by using a vision-attribute semantic connection network according to the obtained image visual feature vector and semantic feature vector to obtain a splicing result; and S5, according to the splicing result, generating a score of each type of the original image in the semantic space through the obtained molecular network, and outputting a final prediction result according to the scores. According to the invention, the problem of image category label missing can be effectively solved.

Description

technical field [0001] The invention relates to a zero-sample image classification method, in particular to a zero-sample image classification method using full data training. Background technique [0002] In the process of image classification, if you want to accurately classify images, you need to inform the model of the image labels of each category. However, the number of image categories is often very large, and new categories may be added from time to time. If each category label is manually marked each time, the workload will be extremely huge. In this process, some categories have only a few or no training sample labels, and the category samples without training labels for the entire category are called zero samples. Such zero samples cannot be effectively constructed by using traditional machine learning methods to construct classifiers. Because traditional models need to use labeled samples to construct a category of parametric / non-parametric models, whether it is...

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): G06K9/62
CPCG06F18/214G06F18/241
Inventor 廖祥文肖永强丘永旺徐戈陈开志
Owner FUZHOU 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