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

Cross-media retrieval method

A cross-media and media technology, applied in the field of pattern recognition and multimedia retrieval, can solve problems such as not being able to adapt to more and more diverse user needs, and achieve the effects of improved accuracy, wide applicability, and good overall performance

Active Publication Date: 2016-11-09
PEKING UNIV SHENZHEN GRADUATE SCHOOL
View PDF3 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Traditional content-based single-media retrieval cannot meet increasingly diverse user needs. For example, in an e-commerce application scenario, a user submits a photo of a "skirt" hoping to find relevant text descriptions or comments; Or find the best matching "dress" photos by submitting some queries

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 retrieval method
  • Cross-media retrieval method
  • Cross-media retrieval method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] Below in conjunction with accompanying drawing, further describe the present invention through embodiment, but do not limit the scope of the present invention in any way.

[0033] A cross-media retrieval method, through coupling-based feature mapping, association mining within the same media, and association mining between different media, to obtain the similarity between images and texts, and perform cross-media retrieval; including the following steps:

[0034] 1) Establish a sample set including image and text media, which is divided into training samples and test samples.

[0035] The sample set of two kinds of media, image and text, both have training samples and test samples respectively.

[0036] 2) Coupling feature mapping is performed on all samples in the sample set. The purpose is to map heterogeneous image and text features into a homogeneous space, and semantic information such as categories should be preserved during the mapping process.

[0037] 3) Coupl...

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 cross-media retrieval method. Cross-media retrieval can be carried out through characteristic mapping based on coupling and association mining in the same media or different media. The cross-media retrieval method comprises the following steps: establishing a multi-media data set; carrying out coupling characteristic mapping on all samples in a sample set; carrying out association mining in the same type of media; carrying out association mining in different types of media; obtaining a class probability matrix of images to be retrieved or texts to be retrieved through two association mining methods and weighting; respectively obtaining a class similarity matrix of image retrieval texts and a class similarity matrix of a text retrieval images, arraying similarities in a descending manner and retrieving. According to the technical scheme provided by the invention, the cross-media retrieval method is good in overall performance and good in applicability, and can be used in various cross-media retrieval scenes.

Description

technical field [0001] The invention relates to the technical fields of pattern recognition and multimedia retrieval, in particular to a cross-media data retrieval method for image and text data based on coupling feature mapping and association mining. Background technique [0002] Since the 1990s, mankind has entered the information society. With the rise of Web 2.0 technology and the widespread use of the Internet, rich multimedia data, such as text, images, audio, and video, have shown explosive growth. Traditional content-based single-media retrieval cannot meet increasingly diverse user needs. For example, in an e-commerce application scenario, a user submits a photo of a "skirt" hoping to find relevant text descriptions or comments; Or find the best matching "dress" photo by submitting a few queries. Therefore, cross-media retrieval technology emerged as the times require, and quickly became a frontier hotspot in the field of multimedia research. Specifically, by onl...

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): G06F17/30
CPCG06F16/00G06F16/43
Inventor 王文敏范梦迪王荣刚李革董胜富王振宇李英赵辉高文
Owner PEKING UNIV SHENZHEN GRADUATE SCHOOL
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