CAKE: Compact and Accurate K-dimensional representation of Emotion

Corentin Kervadec 1 Valentin Vielzeuf 1, 2 Stéphane Pateux 1 Alexis Lechervy 2 Frédéric Jurie 2
2 Equipe Image - Laboratoire GREYC - UMR6072
GREYC - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen
Abstract : Numerous models describing the human emotional states have been built by the psychology community. Alongside, Deep Neural Networks (DNN) are reaching excellent performances and are becoming interesting features extraction tools in many computer vision tasks. Inspired by works from the psychology community, we first study the link between the compact two-dimensional representation of the emotion known as arousal-valence, and discrete emotion classes (e.g. anger, happiness, sadness, etc.) used in the computer vision community. It enables to assess the benefits -- in terms of discrete emotion inference -- of adding an extra dimension to arousal-valence (usually named dominance). Building on these observations, we propose CAKE, a 3-dimensional representation of emotion learned in a multi-domain fashion, achieving accurate emotion recognition on several public datasets. Moreover, we visualize how emotions boundaries are organized inside DNN representations and show that DNNs are implicitly learning arousal-valence-like descriptions of emotions. Finally, we use the CAKE representation to compare the quality of the annotations of different public datasets.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01849908
Contributor : Corentin Kervadec <>
Submitted on : Thursday, August 2, 2018 - 9:22:08 AM
Last modification on : Thursday, February 7, 2019 - 5:45:30 PM
Long-term archiving on : Saturday, November 3, 2018 - 12:55:28 PM

Files

bmvc_review.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01849908, version 2
  • ARXIV : 1807.11215

Citation

Corentin Kervadec, Valentin Vielzeuf, Stéphane Pateux, Alexis Lechervy, Frédéric Jurie. CAKE: Compact and Accurate K-dimensional representation of Emotion. Image Analysis for Human Facial and Activity Recognition (BMVC Workshop), Dr. Zhaojie Ju, Sep 2018, Newcastle, United Kingdom. ⟨hal-01849908v2⟩

Share

Metrics

Record views

218

Files downloads

189