Mutual Information-Based Emotion Recognition

  • Yue Cui
  • Suhuai Luo
  • Qi Tian
  • Shiliang Zhang
  • Yu Peng
  • Lei Jiang
  • Jesse S. Jin
Conference paper

DOI: 10.1007/978-1-4614-3501-3_39

Cite this paper as:
Cui Y. et al. (2013) Mutual Information-Based Emotion Recognition. In: The Era of Interactive Media. Springer, New York, NY

Abstract

Emotions that arise in viewers in response to videos play an essential role in content-based indexing and retrieval. However, the emotional gap between low-level features and high-level semantic meanings is not well understood. This paper proposes a general scheme for video emotion identification using mutual information-based feature selection followed by regression. Continuous arousal and valence values are used to measure video affective content in dimensional arousal-valence space. Firstly, rich audio-visual features are extracted from video clips. The minimum redundancy and maximum relevance feature selection is then used to select most representative feature subsets for arousal and valence modelling. Finally support vector regression is employed to model arousal and valence estimation functions. As evaluated via tenfold cross-validation, the estimation results achieved by our scheme for arousal and valence are: mean absolute error, 0.1358 and 0.1479, variance of absolute error, 0.1074 and 0.1175, respectively. Encouraging results demonstrate the effectiveness of our proposed method.

Key words

Affective content analysis Mutual information-based feature selection Support vector regression 

Copyright information

© Springer Science+Business Media, LLC 2013

Authors and Affiliations

  • Yue Cui
    • 1
  • Suhuai Luo
    • 1
  • Qi Tian
    • 2
  • Shiliang Zhang
    • 3
  • Yu Peng
    • 1
  • Lei Jiang
    • 1
  • Jesse S. Jin
    • 1
  1. 1.School of Design, Communication and Information TechnologyUniversity of NewcastleNewcastle upon TyneAustralia
  2. 2.Department of Computer ScienceUniversity of Texas at San AntonioSan AntonioUSA
  3. 3.Key Laboratory of Intelligent Information Processing, Institute of Computing TechnologyChinese Academy of SciencesBeijingChina

Personalised recommendations