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

Dynamic graph classification method and device based on time domain attention pooling network

A classification method and a classification device technology, which are applied in the direction of instruments, character and pattern recognition, computer components, etc., can solve the problem of reducing the classification performance of moving picture content, and achieve the effect of reducing information interference and improving accuracy

Active Publication Date: 2021-08-06
TIANJIN UNIV +1
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The main challenge of the GIF content classification problem is that some frames in the GIF are irrelevant to the GIF content label, which may greatly degrade the performance of the GIF content classification

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
  • Dynamic graph classification method and device based on time domain attention pooling network
  • Dynamic graph classification method and device based on time domain attention pooling network
  • Dynamic graph classification method and device based on time domain attention pooling network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] Table 1 shows the performance comparison between this method and other methods on the WGIF dataset;

[0039] Table 2 shows the results of the ablation study on the WGIF dataset.

[0040] In order to make the purpose, technical solution and advantages of the present invention clearer, the implementation manners of the present invention will be further described in detail below.

[0041] In the first aspect, an embodiment of the present invention provides a method for classifying animated GIFs based on a time-domain attention pooling network, see figure 1 , the method includes:

[0042] 1. Build WGIF (Web GIF) dataset

[0043] Firstly, a new GIF animation dataset WGIF (a dataset for animated GIF content classification task) with content labels is collected and constructed. In order to ensure the diversity and uniqueness of the constructed data set, the WGIF data set is used to collect typical GIF animations from several web pages, and the animated GIFs in the WGIF data...

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 an animation GIF content classification method and device based on a time domain attention pooling network, and the method comprises the steps: constructing a network dynamic graph data set, and constructing an animation GIF content classification model which comprises a feature extraction module, the time domain attention pooling network and a loss layer; training the content classification model through the constructed network dynamic graph data set, carrying out cross entropy loss evaluation on an output result, adding an auxiliary supervision mode when classifying each frame, and obtaining an overall loss function of the content classification model; and on the basis of the total loss function, capturing a key frame which is most related to the content label in the dynamic picture, and realizing animation classification. The device comprises a construction module, an acquisition module and an animation classification module. The information interference of irrelevant frames is reduced, and the accuracy of animation GIF content classification is improved.

Description

technical field [0001] The present invention relates to the field of moving picture classification, including the construction of moving picture data sets for content classification and the moving picture classification technology based on temporal attention pooling network, in particular to a moving picture classification method based on temporal attention pooling network and devices. Background technique [0002] Animated Graphics Interchange Format (GIF) is an image format with wide compatibility and portability. The GIF image format was created in 1987 by the Internet provider company CompuServe. Contrary to other popular image formats, GIF is better at conveying various forms of emotion, telling stories, and presenting dynamic content. In addition, because there is no sound and long-term information in the GIF image, it is lighter and easier to disperse compared to the video GIF image. With these desirable attributes, the animated GIF format is playing an increasingl...

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): G06K9/62G06K9/46
CPCG06V10/40G06F18/214G06F18/241G06F18/29
Inventor 马永娟朱鹏飞黄进晟王汉石石红赵帅胡清华
Owner TIANJIN UNIV
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