Hierarchical Attention-Based Multimodal Fusion Network for Video Emotion Recognition

Figure 4

Samples with their emotion scores predicted by HAMF.

(a)

(b)

(c)

(d)

(e)

(f)