Skip to main content

Table 3 Best operating multimodal for music video emotion prediction

From: Deep learning-based late fusion of multimodal information for emotion classification of music video

Metrics C3D + 1D Music CNN I3D + 1D Music CNN C3D + 2D Music CNN I3D + 2D Music CNN Integrated multimodal
Test Set Accuracy
(percentage)
Minimum 68.389 66.003 81.709 83.300 86.282
Mean 70.941 69.781 84.956 84.426 88.568
Maximum 72.962 72.166 87.872 85.884 89.860
F-score 0.70 0.69 0.84 0.84 0.88
ROC AUC Score 0.917 0.925 0.979 0.977 0.987