A new model called MAGCAF fuses four pretrained vision systems to recognise student emotions from video, outperforming single ...