Wavelet Sparse Coding-Based Lightweight Networks for Image Superresolution

<table class="table-group" id="tab2"><tr><td><table class="table"><tr><td class="thead-hr" colspan="4"><hr/></td></tr><tr class="thead"><td class="align_left">Activation function</td><td class="align_center">PSNR (float32)</td><td class="align_center">PSNR (uint8)</td><td class="align_center">Loss</td></tr><tr><td class="thead-hr" colspan="4"><hr/></td></tr><tr><td class="align_left">Linear</td><td class="align_center">30.68</td><td class="align_center">24.62</td><td class="align_center">6.06</td></tr><tr><td class="align_left">Clipped ReLU</td><td class="align_center">30.66</td><td class="align_center">29.88</td><td class="align_center">0.78</td></tr><tr class="table-tr"><td colspan="4"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Comparisons of Clipped ReLU and linear activations in the DIV2K [<a href="/journals/misy/2022/2088245/#B35" target="_blank">35</a>] validation dataset.</div>

Mobile Information Systems

tab2

Table 2

Table 2: Wavelet Sparse Coding-Based Lightweight Networks for Image Superresolution 