Irene Kotsia

TL;DR: The paper presents the database description, the experimental set up, the baseline method used for the Challenge, the summary of the performance of the different methods submitted to the Affect-in-the-Wild Challenge for Valence and Arousal estimation and the challenge demonstrates that meticulously designed deep neural networks can achieve very good performance when trained with in- the-wild data.

TL;DR: This paper utilizes GANs to train a very powerful generator of facial texture in UV space and revisits the original 3D Morphable Models (3DMMs) fitting approaches making use of non-linear optimization to find the optimal latent parameters that best reconstruct the test image but under a new perspective.

TL;DR: In this article, an end-to-end deep neural architecture was proposed for predicting continuous emotion dimensions based on visual cues. But the performance of the proposed architecture was not as good as the state-of-the-art.

TL;DR: The way partial occlusion affects human observers when recognizing facial expressions is indicated and conclusions regarding the pairs of facial expressions misclassifications that each type of Occlusion introduces are drawn.

TL;DR: This paper utilizes 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.