Region Dual Attention-Based Video Emotion Recognition

<table class="table-group" id="tab3"><tr><td><table class="table"><tr><td class="thead-hr" colspan="2"><hr/></td></tr><tr class="thead"><td class="align_left">Method</td><td class="align_center">Accuracy (%)</td></tr><tr><td class="thead-hr" colspan="2"><hr/></td></tr><tr><td class="align_left">ResNet-50 + average aggregation</td><td class="align_center">42.22</td></tr><tr><td class="align_left">ResNet-50 + regional location attention mechanism</td><td class="align_center">44.69</td></tr><tr><td class="align_left">ResNet-50 + regional relationship attention mechanism</td><td class="align_center">44.19</td></tr><tr><td class="align_left">ResNet-50 + regional dual attention mechanism</td><td class="align_center">45.68</td></tr><tr class="table-tr"><td colspan="2"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Evaluation of region dual attention mechanism on the HEIV-I data set.</div>

Computational Intelligence and Neuroscience

tab3

Table 3

Table 3: Region Dual Attention-Based Video Emotion Recognition 