Improved U-Net-Like Network for Visual Saliency Detection Based on Pyramid Feature Attention

<div>Scheme of the channel attention module used in this work [<a href="/journals/wcmc/2022/1108462/#B39">39</a>]. <svg height="8.68572pt" id="M31" style="vertical-align:-0.0498209pt" version="1.1" viewbox="-0.0498162 -8.6359 10.0819 8.68572" width="10.0819pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M748 650H522L515 622L546 617C580 611 587 604 565 575C518 513 469 451 419 393C376 474 349 534 330 580C318 609 325 612 361 618L383 622L392 650H151L144 622C214 616 224 612 257 543L360 327C270 218 187 124 159 95C106 40 92 34 26 28L17 0H252L259 28L236 31C189 37 188 47 209 78C249 136 308 210 377 294L478 79C494 44 487 37 449 32L418 28L409 0H673L680 28C596 34 591 39 554 116L436 361C526 469 574 521 604 553C659 612 669 614 739 622L748 650Z"></path></g></svg> and <svg height="11.4859pt" id="M32" style="vertical-align:-0.04979992pt" version="1.1" viewbox="-0.0498162 -11.4361 13.8775 11.4859" width="13.8775pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M748 650H522L515 622L546 617C580 611 587 604 565 575C518 513 469 451 419 393C376 474 349 534 330 580C318 609 325 612 361 618L383 622L392 650H151L144 622C214 616 224 612 257 543L360 327C270 218 187 124 159 95C106 40 92 34 26 28L17 0H252L259 28L236 31C189 37 188 47 209 78C249 136 308 210 377 294L478 79C494 44 487 37 449 32L418 28L409 0H673L680 28C596 34 591 39 554 116L436 361C526 469 574 521 604 553C659 612 669 614 739 622L748 650Z"></path></g><g transform="matrix(.0091,0,0,-0.0091,9.945,-5.741)"><path d="M310 541L304 571C290 586 211 619 185 610L80 76L131 52L310 541Z"></path></g></svg> represent features before and after weighting, respectively.</div>

Wireless Communications and Mobile Computing

fig3

Figure 3

Figure 3: Improved U-Net-Like Network for Visual Saliency Detection Based on Pyramid Feature Attention 