Auto-context modeling using multiple Kernel learning

Huan Song, Jayaraman J. Thiagarajan, Karthikeyan Natesan Ramamurthy, Andreas Spanias

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

In complex visual recognition systems, feature fusion has become crucial to discriminate between a large number of classes. In particular, fusing high-level context information with image appearance models can be effective in object/scene recognition. To this end, we develop an auto-context modeling approach under the RKHS (Reproducing Kernel Hilbert Space) setting, wherein a series of supervised learners are used to approximate the context model. By posing the problem of fusing the context and appearance models using multiple kernel learning, we develop a computationally tractable solution to this challenging problem. Furthermore, we propose to use the marginal probabilities from a kernel SVM classifier to construct the auto-context kernel. In addition to providing better regularization to the learning problem, our approach leads to improved recognition performance in comparison to using only the image features.

Original languageEnglish (US)
Title of host publication2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings
PublisherIEEE Computer Society
Pages1868-1872
Number of pages5
Volume2016-August
ISBN (Electronic)9781467399616
DOIs
StatePublished - Aug 3 2016
Externally publishedYes
Event23rd IEEE International Conference on Image Processing, ICIP 2016 - Phoenix, United States
Duration: Sep 25 2016Sep 28 2016

Other

Other23rd IEEE International Conference on Image Processing, ICIP 2016
CountryUnited States
CityPhoenix
Period9/25/169/28/16

Fingerprint

Hilbert spaces
Object recognition
Classifiers
Fusion reactions

Keywords

  • Feature fusion
  • Image classification
  • Marginalized kernel
  • Multiple kernel learning

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Song, H., Thiagarajan, J. J., Ramamurthy, K. N., & Spanias, A. (2016). Auto-context modeling using multiple Kernel learning. In 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings (Vol. 2016-August, pp. 1868-1872). [7532682] IEEE Computer Society. https://doi.org/10.1109/ICIP.2016.7532682

Auto-context modeling using multiple Kernel learning. / Song, Huan; Thiagarajan, Jayaraman J.; Ramamurthy, Karthikeyan Natesan; Spanias, Andreas.

2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. Vol. 2016-August IEEE Computer Society, 2016. p. 1868-1872 7532682.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Song, H, Thiagarajan, JJ, Ramamurthy, KN & Spanias, A 2016, Auto-context modeling using multiple Kernel learning. in 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. vol. 2016-August, 7532682, IEEE Computer Society, pp. 1868-1872, 23rd IEEE International Conference on Image Processing, ICIP 2016, Phoenix, United States, 9/25/16. https://doi.org/10.1109/ICIP.2016.7532682
Song H, Thiagarajan JJ, Ramamurthy KN, Spanias A. Auto-context modeling using multiple Kernel learning. In 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. Vol. 2016-August. IEEE Computer Society. 2016. p. 1868-1872. 7532682 https://doi.org/10.1109/ICIP.2016.7532682
Song, Huan ; Thiagarajan, Jayaraman J. ; Ramamurthy, Karthikeyan Natesan ; Spanias, Andreas. / Auto-context modeling using multiple Kernel learning. 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. Vol. 2016-August IEEE Computer Society, 2016. pp. 1868-1872
@inproceedings{55085ce4944f481093bebe68851f9251,
title = "Auto-context modeling using multiple Kernel learning",
abstract = "In complex visual recognition systems, feature fusion has become crucial to discriminate between a large number of classes. In particular, fusing high-level context information with image appearance models can be effective in object/scene recognition. To this end, we develop an auto-context modeling approach under the RKHS (Reproducing Kernel Hilbert Space) setting, wherein a series of supervised learners are used to approximate the context model. By posing the problem of fusing the context and appearance models using multiple kernel learning, we develop a computationally tractable solution to this challenging problem. Furthermore, we propose to use the marginal probabilities from a kernel SVM classifier to construct the auto-context kernel. In addition to providing better regularization to the learning problem, our approach leads to improved recognition performance in comparison to using only the image features.",
keywords = "Feature fusion, Image classification, Marginalized kernel, Multiple kernel learning",
author = "Huan Song and Thiagarajan, {Jayaraman J.} and Ramamurthy, {Karthikeyan Natesan} and Andreas Spanias",
year = "2016",
month = "8",
day = "3",
doi = "10.1109/ICIP.2016.7532682",
language = "English (US)",
volume = "2016-August",
pages = "1868--1872",
booktitle = "2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings",
publisher = "IEEE Computer Society",
address = "United States",

}

TY - GEN

T1 - Auto-context modeling using multiple Kernel learning

AU - Song, Huan

AU - Thiagarajan, Jayaraman J.

AU - Ramamurthy, Karthikeyan Natesan

AU - Spanias, Andreas

PY - 2016/8/3

Y1 - 2016/8/3

N2 - In complex visual recognition systems, feature fusion has become crucial to discriminate between a large number of classes. In particular, fusing high-level context information with image appearance models can be effective in object/scene recognition. To this end, we develop an auto-context modeling approach under the RKHS (Reproducing Kernel Hilbert Space) setting, wherein a series of supervised learners are used to approximate the context model. By posing the problem of fusing the context and appearance models using multiple kernel learning, we develop a computationally tractable solution to this challenging problem. Furthermore, we propose to use the marginal probabilities from a kernel SVM classifier to construct the auto-context kernel. In addition to providing better regularization to the learning problem, our approach leads to improved recognition performance in comparison to using only the image features.

AB - In complex visual recognition systems, feature fusion has become crucial to discriminate between a large number of classes. In particular, fusing high-level context information with image appearance models can be effective in object/scene recognition. To this end, we develop an auto-context modeling approach under the RKHS (Reproducing Kernel Hilbert Space) setting, wherein a series of supervised learners are used to approximate the context model. By posing the problem of fusing the context and appearance models using multiple kernel learning, we develop a computationally tractable solution to this challenging problem. Furthermore, we propose to use the marginal probabilities from a kernel SVM classifier to construct the auto-context kernel. In addition to providing better regularization to the learning problem, our approach leads to improved recognition performance in comparison to using only the image features.

KW - Feature fusion

KW - Image classification

KW - Marginalized kernel

KW - Multiple kernel learning

UR - http://www.scopus.com/inward/record.url?scp=85006722228&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006722228&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2016.7532682

DO - 10.1109/ICIP.2016.7532682

M3 - Conference contribution

VL - 2016-August

SP - 1868

EP - 1872

BT - 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings

PB - IEEE Computer Society

ER -