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

Sparse Gaussian process regression method for visual mapping

A Gaussian process regression, vision technology, applied in the field of vision mapping, can solve the problem of no feature selection function, high-dimensional overfitting, etc.

Inactive Publication Date: 2017-05-10
UNIV OF ELECTRONIC SCI & TECH OF CHINA
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing Gaussian process methods for solving visual mapping problems usually have no feature selection function, and too high-dimensional input features often cause overfitting problems.

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
  • Sparse Gaussian process regression method for visual mapping
  • Sparse Gaussian process regression method for visual mapping
  • Sparse Gaussian process regression method for visual mapping

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0069] Implementation language: Matlab, C / C++

[0070] Hardware platform: Intel core2E7400+4G DDR RAM

[0071] Software platform: Matlab2012a, VisualStdio2010

[0072] According to the method of the present invention, first collect a certain number of images and record the corresponding target values ​​of these images, according to the patent of the present invention, utilize Matlab or C++ language to write the sparse Gaussian process regression model program, and train the corresponding model of the present invention on the collected data parameters; then install the camera in various application scenarios to collect the original image, and extract the gradient orientation histogram features; according to the previously trained parameters, the corresponding output target value of the image can be estimated. The method of the present invention can be used for visual mapping problems in various scenarios.

[0073] A sparse Gaussian process regression method for visual mapping...

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 provides a Gaussian process regression method with feature selection performance, and solves the problems of excessively high feature input dimension and feature redundancy in a process of solving a visual mapping problem by utilizing Gaussian process regression. The method comprises the steps of firstly assuming that parameters related to feature weights in distribution parameters meet Laplace priori distribution; secondly estimating the parameters through a maximum posterior probability estimation method, so that feature-related weights have sparsity, namely, feature selection performance; and finally obtaining regression parameters by solving a sparse problem. When a new sample is subjected to visual estimation, sparse regression parameters are directly utilized for performing feature selection and an estimated target value is obtained. The method is simple and effective, and has good performance.

Description

technical field [0001] The invention belongs to the technical field of computer vision, relates to visual mapping technology, and is mainly used in visual estimation problems such as head posture estimation, line of sight tracking and age estimation. Background technique [0002] In computer vision, visual mapping refers to the process of learning a mapping function between input image features and output variables, so that when a new image is input, the target output value corresponding to the input image is estimated. Specifically, visual mapping includes: human body pose estimation, head pose estimation, line of sight estimation, and object tracking. See references for details: O.Williams, A.Blake, and R.Cipolla, Sparse and Semi-Supervised Visual Mapping with the S3GP, in IEEEConference Computer on Computer Vision and Pattern Recognition, pp.230-237, 2006. [0003] As an important branch of computer vision, visual mapping has changed the situation where humans estimate t...

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): G06K9/46
CPCG06V10/50
Inventor 潘力立
Owner UNIV OF ELECTRONIC SCI & TECH OF CHINA
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