A Deep Multimodal Model for Predicting Affective Responses Evoked by Movies Based on Shot Segmentation

Table 3

Comparison of state-of-the-art results for intended emotion prediction.

Models	Arousal		Valence
	MSE	PCC	MSE	PCC
Malandrakis et al. [4]	0.17	0.54	0.24	0.23
Goyal et al. [6]	—	0.62 ± 0.16	—	0.29 ± 0.16
Sivaprasad et al. [7]	0.08 ± 0.04	0.84 ± 0.06	0.21 ± 0.06	0.50 ± 0.14
Thao et al. [9]	0.13	0.62	0.19	0.25
Thao et al. [10]	0.124	0.630	0.178	0.572
Ours (loss1)	0.1022	0.6748	0.1654	0.3167
Ours (loss2)	0.1141	0.6582	0.1704	0.4025