Recognition of affect in the wild using deep neural networks

Conference item


Kollias, D., Nicolaou, M., Kotsia, I., Zhao, G. and Zafeiriou, S. 2017. Recognition of affect in the wild using deep neural networks. CVPRW 2017: 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Honolulu, HI, USA Institute of Electrical and Electronics Engineers (IEEE). pp. 1972-1979 https://doi.org/10.1109/CVPRW.2017.247
TitleRecognition of affect in the wild using deep neural networks
AuthorsKollias, D., Nicolaou, M., Kotsia, I., Zhao, G. and Zafeiriou, S.
Abstract

In this paper we utilize the first large-scale "in-the-wild" (Aff-Wild) database, which is annotated in terms of the valence-arousal dimensions, to train and test an end-to-end deep neural architecture for the estimation of continuous emotion dimensions based on visual cues. The proposed architecture is based on jointly training convolutional (CNN) and recurrent neural network (RNN) layers, thus exploiting both the invariant properties of convolutional features, while also modelling temporal dynamics that arise in human behaviour via the recurrent layers. Various pre-trained networks are used as starting structures which are subsequently appropriately fine-tuned to the Aff-Wild database. Obtained results show premise for the utilization of deep architectures for the visual analysis of human behaviour in terms of continuous emotion dimensions and analysis of different types of affect.

LanguageEnglish
ConferenceCVPRW 2017: 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Page range1972-1979
ISSN2160-7516
ISBN
Electronic9781538607336
Paperback9781538607343
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Publication dates
Print21 Jul 2017
Online24 Aug 2017
Publication process dates
Deposited16 Jun 2017
Accepted06 May 2017
Output statusPublished
Accepted author manuscript
Digital Object Identifier (DOI)https://doi.org/10.1109/CVPRW.2017.247
Book title2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Permalink -

https://repository.mdx.ac.uk/item/87071

Download files

  • 15
    total views
  • 0
    total downloads
  • 2
    views this month
  • 0
    downloads this month

Export as