[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84458-en":3,"doc-seo-84458-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},84458,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1782109480056885918",8,"Research & Report","SwinIFS Landmark-Guided Swin Transformer Identity-Preserving Face Super-Resolution","Face super-resolution reconstructs high-quality facial images from severely degraded low-resolution inputs, yet it remains difficult to recover fine structural details and identity-specific features. SwinIFS presents a landmark-guided super-resolution framework that injects dense Gaussian heatmaps of facial landmarks into a hierarchical attention model. A compact Swin Transformer backbone captures long-range context while preserving local geometry, improving perceptual sharpness and identity retention. Experiments on CelebA show strong results even at 8× magnification, with a favorable balance of reconstruction quality and computational efficiency.","arXiv :2601 .01406v3 [ cs .CV] 11 Jul 2026  \nSwinIFS: Landmark-Guided Swin Transformer For Identity-Preserving Face Super-Resolution  \nHabiba Kauser 1 , Saeed Anwar2 , Omar Hammad 1 , Ibrahim Radwan3 , and Abdul Bais4  \n1 King Fahd University of Petroleum and Minerals, KSA  \n2 The University of Western Australia, Australia  \n3 The University of Canberra, Australia  \n4 The University of Regina, Canada  \nAbstract. Face super-resolution aims to recover high-quality facial images from severely degraded low-resolution inputs, but remains challenging due to the loss of fine structural details and identity-specific features.  \nThis work introduces SwinIFS, a landmark-guided super-resolution framework that integrates structural priors with hierarchical attention mechanisms to achieve identity-preserving reconstruction at both moderate and extreme upscaling factors. The method incorporates dense Gaussian heatmaps of key facial landmarks into the input representation, enabling the network to focus on semantically important facial regions from the earliest stages of processing. A compact Swin Transformer backbone is employed to capture long-range contextual information while preserving local geometry, allowing the model to restore subtle facial texturesand maintain global structural consistency. Extensive experiments on the CelebA benchmark demonstrate that SwinIFS achieves superior perceptual quality, sharper reconstructions, and improved identity retention; it consistently produces more photorealistic results and exhibits strong performance even under 8 × magnification, where most methods fail to recover meaningful structure. SwinIFS also provides an advantageous balance between reconstruction accuracy and computational efficiency, making it suitable for real-world applications in facial enhancement, surveillance, and digital restoration. Our code, model weights, and results are available at [https://github.com/Habiba123-stack/SwinIFS](https://github.com/Habiba123-stack/SwinIFS).  \nKeywords: Face Super-Resolution, Swin Transformer, Identity Preserving SR.  \n1 Introduction  \nFace super-resolution (FSR) aims to reconstruct high-resolution (HR) facial images from low-resolution (LR) inputs while preserving structural coherence and identity-specific details. Reliable recovery of facial features is essential for applications such as surveillance, biometrics, forensics, video conferencing, and media enhancement [12, 28] . Unlike generic super-resolution [1, 5], FSR benefits from the strong geometric regularity of human faces, where the spatial arrangement  \n2 H. Kausar et al.  \nLR SRGAN DIC SISN WIPA W-NET Ours HR  \nFig. 1: A comparison of face super-resolution results is presented, with the top row illustrating results for a 4× upscaling and the bottom row for an 8 × upscaling. Our approach exhibits superior detail recovery and preservation of structure under both degradation conditions.  \nof key components (eyes, nose, mouth) provides valuable prior information for reconstruction. The LR observation process is typically modeled as  \nILR =↓s (IHR ∗ k) + η, (1)  \nwhere IHR is the HR image, ILR is the LR image, k denotes the blur kernel, ↓s is the downsampling operator, and η represents noise. In practical environments, degradation is further compounded by compression artifacts, illumination variations, and sensor noise. At moderate upscaling factors (e.g., 4×), some structural cues remain; however, at extreme scales (e.g. , 8 × inputs), most identity cues are lost, rendering the reconstruction highly ill-posed.  \nEarly face hallucination methods relied on interpolation, example-based patch retrieval, or sparse coding [2] . Although pioneering, these approaches produced overly smooth results and lacked robustness to domain variation. The introduction of deep learning significantly advanced SR performance. CNN-based methods [8, 14 , 17] improved texture reconstruction but remained limited by their local receptive fields, often leading to globa","cbCaiuazRU9ZYo2b","https://ap.wps.com/l/cbCaiuazRU9ZYo2b","pdf",4499264,1,20,"English","en",105,"# Introduction\n## Problem and challenges in face super-resolution\n## Prior methods: hallucination, GANs, and transformers\n## Landmark priors and motivation for SwinIFS\n# Method overview and contributions\n## Landmark-guided multiscale Swin Transformer framework\n## Fusion of RGB appearance and landmark geometry","[{\"question\":\"What problem does SwinIFS address in face super-resolution?\",\"answer\":\"SwinIFS targets identity loss and structural degradation when generating high-resolution faces from severely degraded low-resolution inputs, especially at extreme upscaling factors like 8×.\"},{\"question\":\"How does SwinIFS use facial landmarks?\",\"answer\":\"It encodes key facial landmarks as dense Gaussian heatmaps and feeds them into the input representation to guide the network toward semantically important facial regions and preserve identity-specific structure.\"},{\"question\":\"Why choose a Swin Transformer backbone in SwinIFS?\",\"answer\":\"The compact Swin Transformer captures long-range contextual information through hierarchical window-based attention while maintaining local geometry, supporting consistent global facial structure and subtle texture recovery.\"}]",1784195743,50,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"swinifs-landmark-guided-swin-transformer-identity-preserving-face-super-resolution","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/swinifs-landmark-guided-swin-transformer-identity-preserving-face-super-resolution/84458/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What problem does SwinIFS address in face super-resolution?","Question",{"text":74,"@type":75},"SwinIFS targets identity loss and structural degradation when generating high-resolution faces from severely degraded low-resolution inputs, especially at extreme upscaling factors like 8×.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does SwinIFS use facial landmarks?",{"text":79,"@type":75},"It encodes key facial landmarks as dense Gaussian heatmaps and feeds them into the input representation to guide the network toward semantically important facial regions and preserve identity-specific structure.",{"name":81,"@type":72,"acceptedAnswer":82},"Why choose a Swin Transformer backbone in SwinIFS?",{"text":83,"@type":75},"The compact Swin Transformer captures long-range contextual information through hierarchical window-based attention while maintaining local geometry, supporting consistent global facial structure and subtle texture recovery.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,113,118,121,125,128,132],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":28,"slug":112},6,"Technology","technology",{"id":114,"doc_module":4,"doc_module_name":45,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":119,"slug":120},30,"research-report",{"id":122,"doc_module":4,"doc_module_name":45,"category_name":123,"show_sort_weight":21,"slug":124},9,"Religion & Spirituality","religion-spirituality",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":126,"show_sort_weight":21,"slug":127},"World Cup","world-cup",{"id":129,"doc_module":4,"doc_module_name":45,"category_name":130,"show_sort_weight":129,"slug":131},10,"Lifestyle","lifestyle",{"id":133,"doc_module":4,"doc_module_name":45,"category_name":134,"show_sort_weight":105,"slug":135},19,"General","general"]