The Many Variations of Emotion

Abstract

This paper presents a novel approach to the facial expression generation problem. Building upon the assumption of the psychological community that emotion is intrinsically continuous, we first design our own continuous emotion representation with a 3-dimensional latent space issued from a neural network trained on discrete emotion classification. The so-obtained representation can be used to annotate large in the wild datasets and later used to trained a Generative Adversarial Network.We first show that our model is able to map back to discrete emotion classes with a objectively and subjectively better quality of the images than usual discrete approaches. But also that we are able to pave the larger space of possible facial expressions, generating the many moods of emotion. Moreover, two axis in this space may be found to generate similar expression changes as in traditional continuous representations such as arousalvalence. Finally we show from visual interpretation, that the third remaining dimension is highly related to the well-known dominance dimension from psychology.

Generative Adversarial Networks in Human Emotion Synthesis: A Review

“…In the proposed approach, the generator can transform the non-frontal facial images into frontal ones while the identity and the emotion expression are preserved. Moreover, a recent publication [108] relies on a two-step GAN framework. The first component maps images to a 3D vector space.…”

The Conceptualization and Measurement of Emotion in Machine Learning: A Critical Appraisal and Recommendations from Psychology

“…Thus, the chasm dividing emotion theorists in psychology, according to this narrative, would be whether emotions are better described as discrete categories or as a couple of orthogonal dimensions. Examples of this narrative can be found in (Seyeditabari et al, 2018), (Thanapattheerakul, Mao, Amoranto, & Chan, 2018), (Shu et al, 2018), (Poria, Cambria, Bajpai, & Hussain, 2017), (Gunes et al, 2011), (Kervadec et al, 2018), (Vielzeuf et al, 2018), (Sethu et al, 2019), and (Devillers, Vidrascu, & Lamel, 2005). Even though the discrete versus continuous dichotomy is a common narrative, some authors have offered a more complex account of emotion theory in psychology (e.g., Cowie et al, 2001;Fragopanagos & Taylor, 2005).…”

Generative Adversarial Networks in Human Emotion Synthesis:A Review

“…In the proposed approach, the generator can transform the non-frontal facial images into frontal ones while the identity and the emotion expression are preserved. Moreover, a recent publication (Vielzeuf et al, 2019) relies on a two-step GAN framework. The first component maps images to a 3D vector space.…”