In this work, we describe our method for tackling the valence-arousal estimation challenge from ABAW2 ICCV-2021 Competition. The competition organizers provide an in-the-wild Aff-Wild2 dataset for participants to analyze affective behavior in real-life settings. We use a two stream model to learn emotion features from appearance and action respectively. To solve data imbalanced problem, we apply label distribution smoothing (LDS) to re-weight labels. Our proposed method achieves Concordance Correlation Coefficient (CCC) of 0.591 and 0.617 for valence and arousal on the validation set of Aff-wild2 dataset.
Cite
@article{arxiv.2107.03891,
title = {Technical Report for Valence-Arousal Estimation in ABAW2 Challenge},
author = {Hong-Xia Xie and I-Hsuan Li and Ling Lo and Hong-Han Shuai and Wen-Huang Cheng},
journal= {arXiv preprint arXiv:2107.03891},
year = {2021}
}
Comments
arXiv admin note: substantial text overlap with arXiv:2105.01502