Multimodal Affective Analysis combining Regularized Linear Regression and Boosted Regression Trees

In this paper we present a multimodal approach for affective analysis that exploits features from video, audio, Electrocardiogram (ECG), and Electrodermal Activity (EDA) combining two regression techniques, namely Boosted Regression Trees and Linear Regression. Moreover, we propose a novel regularization approach for the Linear Regression in order to exploit the temporal correlation of the affective dimensions. The final prediction is obtained using a decision level fusion of the regressors individually trained on the different groups of features. The promising results obtained on the benchmark dataset show the efficacy and effectiveness of the proposed approach.