Joint Generative Image Deblurring Aided by Edge Attention Prior and Dynamic Kernel Selection

<table class="figure-group"><tr class="fig-image" id="a"><td><object data="https://static-preview.hindawi.com/articles/wcmc/volume-2021/1391801/figures/1391801.fig.003a.svgz" name="1391801.fig.003a" type="image/svg+xml"></object></td></tr><tr class="fig-caption"><td><b>(a) </b>Fusion</td></tr><tr class="fig-image" id="b"><td><object data="https://static-preview.hindawi.com/articles/wcmc/volume-2021/1391801/figures/1391801.fig.003b.svgz" name="1391801.fig.003b" type="image/svg+xml"></object></td></tr><tr class="fig-caption"><td><b>(b) </b>CRP</td></tr><tr class="fig-image" id="c"><td><object data="https://static-preview.hindawi.com/articles/wcmc/volume-2021/1391801/figures/1391801.fig.003c.svgz" name="1391801.fig.003c" type="image/svg+xml"></object></td></tr><tr class="fig-caption"><td><b>(c) </b>RCU</td></tr><tr class="fig-image" id="d"><td><object data="https://static-preview.hindawi.com/articles/wcmc/volume-2021/1391801/figures/1391801.fig.003d.svgz" name="1391801.fig.003d" type="image/svg+xml"></object></td></tr><tr class="fig-caption"><td><b>(d) </b>MRF</td></tr></table>

<div>Details of convolutionxal layers: (a) fusion unit, (b) improved CRP module, (c) lightweight network structure of RCU, and (d) MRF unit.</div>

Wireless Communications and Mobile Computing

fig3

Figure 3

Figure 3: Joint Generative Image Deblurring Aided by Edge Attention Prior and Dynamic Kernel Selection 