Facial emotion recognition using geometrical features based deep learning techniques

In recent years, intelligent emotion recognition is active research in computer vision to understand the dynamic communication between machines and humans. As a result, automatic allows machine assess acquire human emotional state predict intents based on facial expression. Researchers mainly focus speech features body motions; identifying affect from expressions remains less explored topic. Hence, this paper proposes novel approach for using optimal geometrical landmarks VGG-19s (FCNN). Here, we utilize Haarcascade detect subject face determine distance angle measurements. The entire process classify extracting relevant with normalized measures. experimental analysis shows high accuracy MUG dataset of 94.22% 86.45% GEMEP datasets, respectively.