DS-AOT: Face Image with Occlusion Inpainting via Adaptive Dynamic Gating Boosts Multiscale Self-Attention

DS-AOT: Face Image with Occlusion Inpainting via Adaptive Dynamic Gating Boosts Multiscale Self-Attention | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 19 May 2025 V1 Latest version Share on DS-AOT: Face Image with Occlusion Inpainting via Adaptive Dynamic Gating Boosts Multiscale Self-Attention Authors : Qian Zhang 0000-0003-0610-273X [email protected] , Wuer BAI , Siyang HONG , Lin TENG , and Shuang LIU Authors Info & Affiliations https://doi.org/10.22541/au.174765488.89200679/v1 165 views 86 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Can face image with partially or heavily occlusion be inpainted? In this paper, it is considered as a decision-making issue. An adaptive gating boosts multi-scale self-attention face image inpainting network is proposed to address the issues; it focuses on the task of inpainting large-area occluded face images in complex backgrounds including insufficient fine-grained texture synthesis, inaccurate color inpainting, and semantic dissonance. Multi-level dilated convolution group is constructed to capture local details and long-range contextual information with the help of a dual adaptive gating mechanism which works as: (1) multi-layer convolution and batch normalization to achieve spatially adaptive feature selection, replacing the traditional fixed fusion method of residual connections; (2) multi-scale self attention mechanism explicitly models the global pixel dependency relationship, solving the problems of structural coherence and fine-grained synthesis in large-scale defect inpainting. A large number of experiment results show that this method improves the PSNR and SSIM metrics by an average of 0 . 284dB and 0 . 0042 on the FFHQ dataset, and improves the FID by an average of 8 . 265%. Especially in scenarios with large areas of occlusion ( > 50%), the FID decreases by 3 . 009, significantly improving the quality of facial image inpainting on complex backgrounds. The adaptive gating boosts multi-scale self-attention work strategy can provide reference for other decision-making issues. Supplementary Material File (page2.pdf) Download 2.03 MB Information & Authors Information Version history V1 Version 1 19 May 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords adaptive gating decision making image inpainting self-attention Authors Affiliations Qian Zhang 0000-0003-0610-273X [email protected] Guizhou Minzu University View all articles by this author Wuer BAI Guizhou Minzu University View all articles by this author Siyang HONG Guizhou Minzu University View all articles by this author Lin TENG Guizhou Minzu University View all articles by this author Shuang LIU Guizhou Minzu University View all articles by this author Metrics & Citations Metrics Article Usage 165 views 86 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Qian Zhang, Wuer BAI, Siyang HONG, et al. DS-AOT: Face Image with Occlusion Inpainting via Adaptive Dynamic Gating Boosts Multiscale Self-Attention. Authorea . 19 May 2025. DOI: https://doi.org/10.22541/au.174765488.89200679/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174765488.89200679/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9febf09b1db88e2e',t:'MTc3OTI4NjM4Mw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();