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

Cross-media similarity measure method and search method based on local association graph

A similarity measurement and local correlation technology, applied in the field of cross-media retrieval, can solve problems such as restriction effects, and achieve the effect of accurate and effective similarity measurement, accurate and effective similarity measurement

Active Publication Date: 2016-06-22
PEKING UNIV
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This finer-grained association can only be effectively utilized, which limits the effect of the above-mentioned existing similarity measurement methods

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
  • Cross-media similarity measure method and search method based on local association graph
  • Cross-media similarity measure method and search method based on local association graph
  • Cross-media similarity measure method and search method based on local association graph

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0028] A kind of cross-media similarity measurement method based on the local association graph of the present invention, its process is as follows figure 1 shown, including the following steps:

[0029] (1) Establish a cross-media data set containing multiple media types, and divide the data set into a training set and a test set. Block all the data in the cross-media data set, and retain the overall data while obtaining partial blocks.

[0030] In this embodiment, the multiple media types are text, image and audio. The segmentation methods for these three media are as follows: the image uses a 3×3 grid for fixed blocks; the text uses punctuation marks as separators, and the part between every two punctuation marks is regarded as a local block; audio follows a fixed time Split by length. Within the framework of the method, other ...

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 cross-media similarity measure method and search method based on a local association graph; the method comprises the following steps: building a cross-media data set containing various media types, segmenting all data, thus obtaining local segments while reserving the whole data; extracting characteristic vectors from whole data and local segments of each media type in the cross-media data set; using a cross-media unification characteristic expression method to train various media types through the whole data in a training set so as to obtain a unification characteristic mapping matrix, and mapping the whole data and local segments of each media type to the unification space; building the local association graph in the unification space, and calculating media data similarity. The cross-media similarity measure method and search method can measure cross-media data similarity, can use local information to dig fine grain cross-media association, can use whole information to effectively block local information noises, thus optimizing cross-media similarity measure exhaustive and accuracy.

Description

technical field [0001] The invention relates to the field of multimedia retrieval, in particular to a cross-media similarity measurement method and a cross-media retrieval method based on a partial association graph. Background technique [0002] With the rapid development of multimedia technology and Internet technology, the image, text, video, audio and other multimedia data on the network are increasing rapidly, and there are more and more multimedia related applications. Users need more flexible and effective information retrieval technology. The traditional information retrieval technology is mainly text-based retrieval, that is, manually annotating the data, and returning the retrieval results by matching the user input text. This retrieval method is time-consuming and labor-intensive, and cannot cope with large-scale multimedia data. Therefore, many scholars have proposed content-based multimedia retrieval, which returns relevant retrieval results according to the si...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/30
CPCG06F16/435
Inventor 彭宇新黄鑫
Owner PEKING 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