Feature regression for multimodal image analysis

Michael Ying Yang*, Xuanzi Yong, Bodo Rosenhahn

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

2 Citations (Scopus)

Abstract

In this paper, we analyze the relationship between the corresponding descriptors computed from multimodal images with focus on visual and infrared images. First the descriptors are regressed by means of linear regression as well as Gaussian process. We apply different covariance functions and inference methods for Gaussian process. Then the descriptors detected from visual images are mapped to infrared images through the regression results. Predictions are assessed in two ways: the statistics of absolute error between true values and actual values, and the precision score of matching the predicted descriptors to the original infrared descriptors. Experimental results show that regression methods achieve a well-assessed relationship between corresponding descriptors from multiple modalities.

Original languageEnglish
Title of host publicationIEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2014
Place of PublicationColumbus
PublisherIEEE Computer Society
Pages770-777
Number of pages8
ISBN (Electronic)9781479943098, 9781479943098
DOIs
Publication statusPublished - 24 Sep 2014
Externally publishedYes
Event27th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014 - Columbus, OH, USA, Columbus, United States
Duration: 23 Jun 201428 Jun 2014
Conference number: 27

Publication series

NameIEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
ISSN (Print)2160-7508
ISSN (Electronic)2160-7516

Conference

Conference27th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014
Abbreviated titleCVPR 2014
CountryUnited States
CityColumbus
Period23/06/1428/06/14
Other23-28 June 2014

Fingerprint

Dive into the research topics of 'Feature regression for multimodal image analysis'. Together they form a unique fingerprint.

Cite this