Micro-video popularity prediction method based on low-rank constraint and multi-view characteristic fusion

A feature fusion, low-rank constraint technology, applied in prediction, character and pattern recognition, special data processing applications, etc., can solve the problems of micro-video pollution, inability to meet, etc., to achieve high stability and improve accuracy.

Inactive Publication Date: 2017-10-03
TIANJIN UNIV
View PDF3 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In practical applications, due to changes in the external environment and camera shakes, micro-videos are polluted, and the features extracted from the video cannot be com

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
  • Micro-video popularity prediction method based on low-rank constraint and multi-view characteristic fusion
  • Micro-video popularity prediction method based on low-rank constraint and multi-view characteristic fusion
  • Micro-video popularity prediction method based on low-rank constraint and multi-view characteristic fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] In order to achieve a better prediction effect, a comprehensive, automatic and accurate method for predicting the popularity of micro-videos is needed. Research shows that micro-videos with similar features have similar popularity. The embodiment of the present invention proposes a micro-video popularity prediction method based on low-rank constraints and multi-view feature fusion, see figure 1 , see the description below:

[0027] 101: Perform low-rank approximation processing on the 4 kinds of viewing angle modal features respectively, and obtain 4 kinds of low-rank feature information for removing noise;

[0028] 102: Feature fusion of 4 kinds of low-rank feature information through canonical correlation analysis of multi-view information;

[0029] 103: Use the fused feature information to establish a Laplacian matrix representing the graph relationship between micro-videos; based on the Laplacian matrix, use a semi-supervised method to predict the popularity of mi...

Embodiment 2

[0039] The scheme in embodiment 1 is further introduced below in conjunction with specific calculation formulas and examples, see the following description for details:

[0040] 201: Extract 4 kinds of view mode features for a given micro-video, namely: visual features, acoustic features, text features and social attribute features;

[0041] In the embodiment of the present invention, four commonly used features of micro-video research are firstly extracted from a given micro-video, including: visual features, acoustic features, text features and social attribute features.

[0042] 1. Visual features include: color histogram information, object information in the micro-video (which can be obtained by convolutional neural networks, or obtained by other methods, which are not limited in the embodiments of the present invention) and aesthetic features.

[0043] 2. Acoustic features include: the music in the micro-video and the features of other main background sounds.

[0044] 3...

Embodiment 3

[0072] The scheme in embodiment 1 and 2 is carried out feasibility verification below in conjunction with specific example, see the following description for details:

[0073] 1. Test data set

[0074] The test data set used in this experiment is a collection of micro-videos downloaded from the Vine social networking site, and the length of the micro-videos is 6S.

[0075] 2. Evaluation criteria

[0076] The mean square error (nMSE) represents the accuracy of the prediction, and the p value (P-value) represents the reliability of the prediction.

[0077] 3. Comparison algorithm

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 micro-video popularity prediction method based on low-rank constraint and multi-view characteristic fusion. The micro-video popularity prediction method includes the steps that four view modal characteristics are subjected to low-rank approximation processing respectively, and four kinds of noise-removing low-rank characteristic information are obtained; through multi-view-information typical correlation analysis, the four kinds of low-rank characteristic information are subjected to characteristic fusion; fused characteristic information is adopted, laplacian matrixes for expressing graph relationships between all micro videos are established; based on the laplacian matrixes, the popularity of the micro videos is predicted with the semi-supervised method. According to the micro-video popularity prediction method based on the low-rank constraint and multi-view characteristic fusion, the limitation of single-view characteristic on popularity prediction is avoided, the characteristics of all the views are processed through low-rank approximation processing, and the laplacian matrixes established between the characteristics have the higher stability.

Description

technical field [0001] The invention relates to the field of micro-video popularity prediction, in particular to a micro-video popularity prediction method based on low-rank constraints and multi-view feature fusion. Background technique [0002] With the popularity of network technology and social platforms, micro-video has received more and more attention as a new user content. Micro-videos refer to short video clips ranging from 30 seconds to 20 minutes in length. The emergence of micro-videos not only conforms to the online viewing habits and mobile terminal characteristics under the fast-paced lifestyle of modern society, but also meets the needs of consumers' independent participation and return on attention in the era of entertainment explosion and attention scarcity. It is foreseeable that What "micro video" brings to the public will be free video enjoyment anytime, anywhere. The prediction of micro-video popularity has a guiding role in advertising push, video rec...

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): G06F17/30G06K9/62G06Q10/04G06Q50/00
CPCG06F16/735G06F16/7834G06F16/7844G06F16/7867G06Q10/04G06Q50/01G06F18/253
Inventor 苏育挺白须井佩光张静
Owner TIANJIN 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