A Joint Cross-Attention Model for Audio-Visual Fusion in Dimensional Emotion Recognition | IEEE Conference Publication | IEEE Xplore