The Many Moods of Emotion
This work addresses the challenge of generating realistic and diverse facial expressions for applications in psychology and AI, representing an incremental improvement over existing discrete methods.
The paper tackles the facial expression generation problem by introducing a continuous emotion representation derived from a neural network trained on discrete emotion classification, which is then used to train a Generative Adversarial Network to generate high-quality images that map back to discrete emotions and explore a broader space of facial expressions.
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 arousal-valence. Finally we show from visual interpretation, that the third remaining dimension is highly related to the well-known dominance dimension from psychology.