Affective abstract image classification and retrieval using multiple kernel learning

He Zhang, Zhirong Yang, Mehmet Gonen, Markus Koskela, Jorma Laaksonen, Timo Honkela, Erkki Oja

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

9 Citations (Scopus)

Abstract

Emotional semantic image retrieval systems aim at incorporating the user's affective states for responding adequately to the user's interests. One challenge is to select features specific to image affect detection. Another challenge is to build effective learning models or classifiers to bridge the so-called "affective gap". In this work, we study the affective classification and retrieval of abstract images by applying multiple kernel learning framework. An image can be represented by different feature spaces and multiple kernel learning can utilize all these feature representations simultaneously (i.e., multiview learning), such that it jointly learns the feature representation weights and corresponding classifier in an intelligent manner. Our experimental results on two abstract image datasets demonstrate the advantage of the multiple kernel learning framework for image affect detection in terms of feature selection, classification performance, and interpretation.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages166-175
Number of pages10
Volume8228 LNCS
EditionPART 3
DOIs
StatePublished - 2013
Externally publishedYes
Event20th International Conference on Neural Information Processing, ICONIP 2013 - Daegu, Korea, Republic of
Duration: Nov 3 2013Nov 7 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 3
Volume8228 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other20th International Conference on Neural Information Processing, ICONIP 2013
CountryKorea, Republic of
CityDaegu
Period11/3/1311/7/13

Fingerprint

Image classification
Image Classification
Image retrieval
Image Retrieval
Classifiers
kernel
Feature extraction
Semantics
Classifier
Feature Space
Feature Selection
Retrieval
Learning
Experimental Results
Demonstrate
Framework

Keywords

  • Group lasso
  • Image affect
  • Image classification and retrieval
  • Low-level image features
  • Multiple kernel learning

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Zhang, H., Yang, Z., Gonen, M., Koskela, M., Laaksonen, J., Honkela, T., & Oja, E. (2013). Affective abstract image classification and retrieval using multiple kernel learning. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (PART 3 ed., Vol. 8228 LNCS, pp. 166-175). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8228 LNCS, No. PART 3). https://doi.org/10.1007/978-3-642-42051-1_22

Affective abstract image classification and retrieval using multiple kernel learning. / Zhang, He; Yang, Zhirong; Gonen, Mehmet; Koskela, Markus; Laaksonen, Jorma; Honkela, Timo; Oja, Erkki.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8228 LNCS PART 3. ed. 2013. p. 166-175 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8228 LNCS, No. PART 3).

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

Zhang, H, Yang, Z, Gonen, M, Koskela, M, Laaksonen, J, Honkela, T & Oja, E 2013, Affective abstract image classification and retrieval using multiple kernel learning. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 3 edn, vol. 8228 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 3, vol. 8228 LNCS, pp. 166-175, 20th International Conference on Neural Information Processing, ICONIP 2013, Daegu, Korea, Republic of, 11/3/13. https://doi.org/10.1007/978-3-642-42051-1_22
Zhang H, Yang Z, Gonen M, Koskela M, Laaksonen J, Honkela T et al. Affective abstract image classification and retrieval using multiple kernel learning. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 3 ed. Vol. 8228 LNCS. 2013. p. 166-175. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 3). https://doi.org/10.1007/978-3-642-42051-1_22
Zhang, He ; Yang, Zhirong ; Gonen, Mehmet ; Koskela, Markus ; Laaksonen, Jorma ; Honkela, Timo ; Oja, Erkki. / Affective abstract image classification and retrieval using multiple kernel learning. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8228 LNCS PART 3. ed. 2013. pp. 166-175 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 3).
@inproceedings{13cb02fbc3f24a108a68e900d3da4730,
title = "Affective abstract image classification and retrieval using multiple kernel learning",
abstract = "Emotional semantic image retrieval systems aim at incorporating the user's affective states for responding adequately to the user's interests. One challenge is to select features specific to image affect detection. Another challenge is to build effective learning models or classifiers to bridge the so-called {"}affective gap{"}. In this work, we study the affective classification and retrieval of abstract images by applying multiple kernel learning framework. An image can be represented by different feature spaces and multiple kernel learning can utilize all these feature representations simultaneously (i.e., multiview learning), such that it jointly learns the feature representation weights and corresponding classifier in an intelligent manner. Our experimental results on two abstract image datasets demonstrate the advantage of the multiple kernel learning framework for image affect detection in terms of feature selection, classification performance, and interpretation.",
keywords = "Group lasso, Image affect, Image classification and retrieval, Low-level image features, Multiple kernel learning",
author = "He Zhang and Zhirong Yang and Mehmet Gonen and Markus Koskela and Jorma Laaksonen and Timo Honkela and Erkki Oja",
year = "2013",
doi = "10.1007/978-3-642-42051-1_22",
language = "English (US)",
isbn = "9783642420504",
volume = "8228 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
number = "PART 3",
pages = "166--175",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
edition = "PART 3",

}

TY - GEN

T1 - Affective abstract image classification and retrieval using multiple kernel learning

AU - Zhang, He

AU - Yang, Zhirong

AU - Gonen, Mehmet

AU - Koskela, Markus

AU - Laaksonen, Jorma

AU - Honkela, Timo

AU - Oja, Erkki

PY - 2013

Y1 - 2013

N2 - Emotional semantic image retrieval systems aim at incorporating the user's affective states for responding adequately to the user's interests. One challenge is to select features specific to image affect detection. Another challenge is to build effective learning models or classifiers to bridge the so-called "affective gap". In this work, we study the affective classification and retrieval of abstract images by applying multiple kernel learning framework. An image can be represented by different feature spaces and multiple kernel learning can utilize all these feature representations simultaneously (i.e., multiview learning), such that it jointly learns the feature representation weights and corresponding classifier in an intelligent manner. Our experimental results on two abstract image datasets demonstrate the advantage of the multiple kernel learning framework for image affect detection in terms of feature selection, classification performance, and interpretation.

AB - Emotional semantic image retrieval systems aim at incorporating the user's affective states for responding adequately to the user's interests. One challenge is to select features specific to image affect detection. Another challenge is to build effective learning models or classifiers to bridge the so-called "affective gap". In this work, we study the affective classification and retrieval of abstract images by applying multiple kernel learning framework. An image can be represented by different feature spaces and multiple kernel learning can utilize all these feature representations simultaneously (i.e., multiview learning), such that it jointly learns the feature representation weights and corresponding classifier in an intelligent manner. Our experimental results on two abstract image datasets demonstrate the advantage of the multiple kernel learning framework for image affect detection in terms of feature selection, classification performance, and interpretation.

KW - Group lasso

KW - Image affect

KW - Image classification and retrieval

KW - Low-level image features

KW - Multiple kernel learning

UR - http://www.scopus.com/inward/record.url?scp=84893413775&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893413775&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-42051-1_22

DO - 10.1007/978-3-642-42051-1_22

M3 - Conference contribution

AN - SCOPUS:84893413775

SN - 9783642420504

VL - 8228 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 166

EP - 175

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -