Automatic Image Captioning Based on ResNet50 and LSTM with Soft Attention

<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">Model</td><td class="align_center">Right choosing of generated description</td></tr><tr><td class="thead-hr" colspan="2"><hr/></td></tr><tr><td class="align_left">With attention</td><td class="align_center">71%</td></tr><tr><td class="align_left">Without attention</td><td class="align_center">54%</td></tr><tr class="table-tr"><td colspan="2"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Comparison for AICRL with and without attention.</div>

Wireless Communications and Mobile Computing

tab3

Table 3

Table 3: Automatic Image Captioning Based on ResNet50 and LSTM with Soft Attention 