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

Query-style image feature representation generation, image search method and device

A picture feature and query technology, applied in the field of information processing, can solve the problems of lack of statistical significance, sparse clicks, and not counting clicks, and achieve the effect of optimizing picture search technology and improving relevance.

Active Publication Date: 2018-04-13
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The above three methods for describing the relevance characteristics of image search all have certain limitations:
[0007] Click-through rate characteristics: mainly based on user behavior statistics. On the one hand, there are biases and noises, and on the other hand, there is sparsity. Only pictures that are displayed at the top and have a sufficient number of times under high-frequency queries can count sufficient clicks. In other cases , no clicks are counted, or the clicks are very sparse, lacking statistical significance

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
  • Query-style image feature representation generation, image search method and device
  • Query-style image feature representation generation, image search method and device
  • Query-style image feature representation generation, image search method and device

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0041] figure 1 It is a flow chart of a method for generating a query-type image feature representation provided in the first embodiment of the present invention. The method of this embodiment can be executed by a device for generating a query-type image feature representation, which can be implemented by hardware and / or or software, and generally can be integrated into a server for generating query-style image feature representations. The method of this embodiment specifically includes:

[0042] 110. Obtain a set of clicked pictures corresponding to the target query according to the picture click log of the search user.

[0043] Generally speaking, after a user enters a query expression through a picture search engine, he or she will click to view one or more pictures based on the picture search results corresponding to the query formula. Correspondingly, the search engine will store the query formula entered by the user and the picture information clicked based on the quer...

no. 2 example

[0063] figure 2 It is a flowchart of a method for generating a query-based image feature representation according to the second embodiment of the present invention. This embodiment is optimized based on the above embodiments. In this embodiment, the feature vector of the picture is specifically optimized as: category distribution vector;

[0064] Correspondingly, according to the picture content features of the picture, generating the picture feature vector corresponding to each clicked picture in the clicked picture set is specifically optimized as follows: input the picture pixel content of the clicked picture to be processed into the category classifier group; wherein, The category classifier group is generated by training with a set number of training data that has been classified and labeled; the first vector output by the category classifier group is obtained, which is composed of the probability values ​​that the clicked pictures to be processed belong to different cat...

no. 3 example

[0082] image 3 It is a flow chart of a method for generating query-based image feature representations according to the third embodiment of the present invention. This embodiment is optimized based on the above embodiments. In this embodiment, each of the clicked pictures will be clustered according to the picture feature vector, and the clustering result will be used as the picture feature of the target query formula. Representing that the specific optimization is as follows: according to the picture feature vector corresponding to each clicked picture and the clustering algorithm matched with the picture feature vector, each of the clicked pictures is clustered to generate a cluster cluster set; The abnormal cluster clusters included in the cluster set; the cluster cluster set after filtering out the abnormal cluster clusters is used as the picture feature representation of the target query expression.

[0083] Correspondingly, the method in this embodiment specifically in...

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 embodiment of the invention discloses a method and device for generating query-type image feature representation and image search. The generation method of the picture feature representation of the query formula includes: according to the picture click log of the search user, obtaining a set of clicked pictures corresponding to the target query formula; Corresponding picture feature vectors: performing clustering on each of the clicked pictures according to the picture feature vectors, and using the clustering result as the picture feature representation of the target query formula. The technical scheme of the invention optimizes the existing picture search technology, and can greatly improve the correlation between the picture search results and the query formula input by the user.

Description

technical field [0001] Embodiments of the present invention relate to information processing technologies, and in particular, to a method and device for generating query-style image feature representations and image search. Background technique [0002] Image search refers to the user's input of natural language queries, for example, through the query formula (also called Query) entered in the text input box provided by the search engine, searching from the image collection and returning the sorted image results according to the relevance and other indicators Information retrieval process for users. Relevance is one of the most important performance indicators of search engines, which measures the degree of relevance between returned results and user query requirements. For an image search engine, the returned image is in an unstructured pixel format, while the query entered by the user is in a text format. These are two completely different information formats, and the two...

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 Patents(China)
IPC IPC(8): G06F17/30G06V10/762
CPCG06F16/583G06V10/762G06F18/23G06F18/24G06F16/5866G06F18/22G06F18/214G06F18/2321G06F18/2411
Inventor 付立波方高林陈雨
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
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