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

Multi-view learning method of joint low-rank representation (LRR) and sparse regression

A low-rank representation, sparse regression technology, applied in character and pattern recognition, instruments, computer parts, etc., to achieve the effect of improving accuracy

Active Publication Date: 2018-01-05
TIANJIN UNIV
View PDF3 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] Using the multi-vision adaptive regression algorithm to solve the problem of automatically predicting the memory of the image, the relationship between the underlying image features, image attribute features and image memory is obtained under the optimal parameters;

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
  • Multi-view learning method of joint low-rank representation (LRR) and sparse regression
  • Multi-view learning method of joint low-rank representation (LRR) and sparse regression
  • Multi-view learning method of joint low-rank representation (LRR) and sparse regression

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] Studies have shown that image attribute features are very high-level semantic features compared to their original underlying features. To study the visual features of images and predict image memory, the embodiment of the present invention proposes a method for predicting image memory A multi-view learning method that combines low-rank representation and sparse regression, see figure 1 , the method includes the following steps:

[0032] 101: Obtain an image memory data set;

[0033] Among them, the image memorability dataset [1] Contains datasets from SUN [11] 2,222 images of . The memory score of the image is obtained through Amazon Mechanical Turk's Visual Memory Game, and the image memory is a continuous value from 0 to 1. The higher the value, the harder the image is to remember. Sample images with various memory scores such as figure 2 shown.

[0034] 102: Extract low-level features and high-level attribute features from the SUN dataset with image memory sc...

Embodiment 2

[0042] The scheme in embodiment 1 is further introduced below in combination with specific calculation formulas, see the following description for details:

[0043] 201: The Image Memorability Dataset contains 2,222 images from the SUN dataset;

[0044] Wherein, the data set is well known to those skilled in the art, and will not be described in detail in this embodiment of the present invention.

[0045]202: Perform feature extraction on the pictures of the SUN dataset with the image memory score label, the extracted SIFT, Gist, HOG and SSIM features constitute a low-level feature library, and use two types of high-level attribute features, including 327-dimensional scene categories Attributes, 106-dimensional object attributes

[0046] Among them, the above data set includes 2222 pictures in various environments, and each picture is marked with an image memory score, attached figure 2 Shows a sample of images from the database labeled with memory scores. Features express...

Embodiment 3

[0107] Combined with the specific experimental data, Figure 3 to Figure 4 The scheme in embodiment 1 and 2 is carried out feasibility verification, see the following description for details:

[0108] The Image Memorability dataset contains 2,222 images from the SUN dataset. The memory score of the image is obtained through Amazon Mechanical Turk's Visual Memory Game, and the image memory is a continuous value from 0 to 1. The higher the value, the harder the image is to remember, sample images with various memory scores such as figure 2 shown.

[0109] This method adopts two evaluation methods:

[0110] Ranking Correlation Evaluation Method (Ranking Correlation, RC): Get the ranking relationship between the real memory ranking and the predicted memory score, and use the ranking-related Spearman correlation coefficient standard to measure the correlation coefficient between the two rankings. Its value range is [-1,1], and the higher the value, the closer the two sorts are...

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 multi-view learning method of joint low-rank representation (LRR) and sparse regression. The method comprises the following steps: respectively carrying out extraction of low-level features and high-level attribute features on an SUN data set with image memory degree score labels; placing three parts of low-rank representation, a sparse regression model and multi-view consistency losses under the same framework to form a whole, and constructing a multi-view-joint low-rank and sparse-regression (Mv-JLRSR) model; utilizing a multi-vision adaptive regression (MAR) algorithm to solve the problem of automatically predicting memoryability of images, and obtaining relationships of the low-layer image features, the image attribute features and image memory degrees under optimal parameters; and combining the low-level features and the high-level attribute features of the images, utilizing relationship results, which are obtained under the optimal parameters, to predictmemory degrees of the images of a database test set, and using relevant evaluation criteria to verify prediction results. According to the multi-view learning framework of joint low-rank representation and sparse regression of the invention, the memoryability of image areas is accurately predicted.

Description

technical field [0001] The invention relates to the field of low-rank representation and sparse regression, in particular to a multi-view learning method combining low-rank representation and sparse regression. Background technique [0002] Humans have the ability to remember thousands of images, however not all images are stored in the brain in the same way. Some representative pictures are remembered at a glance, while other images are easily lost from memory. Image memory is used to measure how well images are remembered or forgotten after a specific period of time. Previous work has shown that memory for pictures is related to intrinsic properties of pictures, namely that memory for pictures is consistent across time intervals and across observers. In this context, just like studying many other high-level image attributes such as popularity, interest, emotion, and aesthetics, several research works started to explore the potential correlation between image content repr...

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): G06K9/62
Inventor 刘安安史英迪苏育挺
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