Channel-Wise Spatiotemporal Aggregation Technology for Face Video Forensics

<table class="table-group" id="tab8"><tr><td><table class="table"><tr><td class="thead-hr" colspan="2"><hr/></td></tr><tr class="thead"><td class="align_left">Approach</td><td class="align_center">Compressed NT</td></tr><tr><td class="thead-hr" colspan="2"><hr/></td></tr><tr><td class="align_left">CWSA</td><td class="align_center"><b>80.6</b></td></tr><tr><td class="align_left">Simple fusion</td><td class="align_center">79.4</td></tr><tr><td class="align_left">LSTM</td><td class="align_center">76.0</td></tr><tr><td class="align_left">RNN</td><td class="align_center">75.6</td></tr><tr class="table-tr"><td colspan="2"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Binary classification accuracy (%) (higher is better) comparison based on the same backbone on highly compressed NT datasets.</div>

Security and Communication Networks

tab8

Table 8

Table 8: Channel-Wise Spatiotemporal Aggregation Technology for Face Video Forensics 