@article{Lee_Wong_2020, title={STeW: Real-time Video Facial Emotion Classification via a Compact Sliding Temporal Windowed Deep Neural Network}, volume={5}, url={https://openjournals.uwaterloo.ca/index.php/vsl/article/view/1667}, abstractNote={<p>The real-time classification of human facial expressions presents a<br>challenging task, even for humans. Individuals with Autism Spectrum Disorder (ASD) have an even greater difficulty in detecting<br>and interpreting these facial expressions, which can lead to an<br>increased risk of depression and loneliness due to a disconnect<br>with society. This study explores a compact Sliding Temporal Windowed (STeW) deep neural network architecture for real-time video<br>facial emotion classification. The proposed STeW architecture is<br>designed to provide a balance between speed and the leveraging<br>of temporal characteristics to capture transient nuances of facial<br>expressions. A more difficult dataset (which we call BigFaceX) is<br>proposed by combining and modifying the extended Cohn-Kanade<br>(CK+), BAUM-1, and the eNTERFACE public datasets, and used to<br>evaluate the proposed STeW network. Experimental results show<br>that the proposed STeW network architecture can achieve noticeably higher accuracy when compared to the highly compact mini-<br>Xception network, thus illustrating the potential for leveraging this<br>approach to achieve real-time video facial emotion classification.</p>}, number={1}, journal={Journal of Computational Vision and Imaging Systems}, author={Lee, James and Wong, Alexander}, year={2020}, month={Jan.}, pages={1} }