{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T12:22:55Z","timestamp":1775737375290,"version":"3.50.1"},"reference-count":63,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,7,11]],"date-time":"2025-07-11T00:00:00Z","timestamp":1752192000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Addressing inpainting challenges in high-resolution images remains a complex task. The most recent image inpainting techniques rely on machine learning models; however, a major limitation of supervised methods is their dependence on end-to-end training. Even minor changes to the input often necessitate retraining, making the process inefficient. As a result, unsupervised learning approaches have gained prominence in image inpainting. State-of-the-art methods, particularly those using generative adversarial networks (GANs), have achieved promising results. However, generating photorealistic outputs for high-resolution images with arbitrary large-region masks remains difficult. Inpainted images often suffer from deformed structures and blurry textures, compromising quality. Additionally, building a model capable of handling a diverse range of images presents further challenges. These challenges are addressed by proposing a novel probabilistic model that utilizes picture priors to learn prominent features within StyleGAN3. The priors are constructed using cosine similarity, mean, and intensity, where intensity is computed using the improved Papoulis\u2013Gerchberg algorithm. The image is reconstructed using the probabilistic maximum <jats:italic>a posteriori<\/jats:italic> estimate. Variational inference is then applied to obtain the optimal solution using a modified Bayes-by-Backprop approach. The model is evaluated on 70,000 images from the Flickr-Faces-HQ, DIV2K, and brain datasets and surpasses state-of-the-art techniques in reconstruction quality.<\/jats:p>","DOI":"10.3389\/frai.2025.1614608","type":"journal-article","created":{"date-parts":[[2025,7,11]],"date-time":"2025-07-11T08:22:41Z","timestamp":1752222161000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["High-resolution image inpainting using a probabilistic framework for diverse images with large arbitrary masks"],"prefix":"10.3389","volume":"8","author":[{"given":"G.","family":"Sumathi","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"M.","family":"Uma Devi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,7,11]]},"reference":[{"key":"B1","first-page":"8296","article-title":"\u201cImage2stylegan++: how to edit the embedded images?.\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Abdal","year":"2020"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1811.05910","article-title":"Deep Bayesian inversion","author":"Adler","year":"2018","journal-title":"arXiv [preprint]. arXiv:1811.05910"},{"key":"B3","unstructured":"\u201cVIPaint: image inpainting with pre-trained diffusion models via variational inference,\u201d\n          \n          \n            \n              Agarwal\n              S.\n            \n            \n              Hope\n              G.\n            \n            \n              Sudderth\n              E. B.\n            \n          \n          The Thirteenth International Conference on Learning Representations, Singapore.\n          \n          2024"},{"key":"B4","doi-asserted-by":"publisher","first-page":"2459","DOI":"10.1007\/s11263-020-01310-5","article-title":"Mimicgan: robust projection onto image manifolds with corruption mimicking","volume":"128","author":"Anirudh","year":"2020","journal-title":"Int. J. Comput. Vis."},{"key":"B5","first-page":"399","article-title":"\u201cInvertible generative models for inverse problems: mitigating representation error and dataset bias,\u201d","author":"Asim","year":"2020","journal-title":"International Conference on Machine Learning"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1145\/3306346.3323023","article-title":"Semantic photo manipulation with a generative image prior","author":"Bau","year":"2020","journal-title":"arXiv [preprint]. arXiv:2005.07727"},{"key":"B7","article-title":"\u201cAmbientGAN: generative models from lossy measurements,\u201d","volume-title":"International Conference on Learning Representations","author":"Bora","year":"2018"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2012.02381","article-title":"Generator pyramid for high-resolution image inpainting","author":"Cao","year":"2020","journal-title":"arXiv [preprint]. arXiv:2012.02381"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.5555\/3327757.3327764","article-title":"Neural ordinary differential equations","author":"Chen","year":"2018","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B10","article-title":"\u201cFlexible, high performance convolutional neural networks for image classification,\u201d","volume-title":"Twenty-Second International Joint Conference on Artificial Intelligence","author":"Ciresan","year":"2011"},{"key":"B11","doi-asserted-by":"publisher","first-page":"672","DOI":"10.1017\/S1041610209009405","article-title":"The Australian Imaging, Biomarkers and Lifestyle (AIBL) study of aging: methodology and baseline characteristics of 1112 individuals recruited for a longitudinal study of Alzheimer's disease","volume":"21","author":"Ellis","year":"2009","journal-title":"Int. Psychogeriatr."},{"key":"B12","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/ICEE50131.2020.9260846","article-title":"\u201cImage inpainting by multiscale spline interpolation,\u201d","volume-title":"2020 28th Iranian Conference on Electrical Engineering (ICEE)","author":"Ghorbanzade","year":"2020"},{"key":"B13","doi-asserted-by":"publisher","first-page":"328","DOI":"10.1109\/TCI.2019.2948732","article-title":"Neumann networks for linear inverse problems in imaging","volume":"6","author":"Gilton","year":"2019","journal-title":"IEEE Trans. Comput. Imaging"},{"key":"B14","doi-asserted-by":"publisher","first-page":"5468","DOI":"10.1109\/TNNLS.2020.2968289","article-title":"Learning deep gradient descent optimization for image deconvolution","volume":"31","author":"Gong","year":"2020","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"B15","first-page":"2672","author":"Goodfellow","year":"2014","journal-title":"Generative Adversarial Nets in Advances in Neural Information Processing Systems (NIPS)"},{"key":"B16","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1145\/1276377.1276382","article-title":"Scene completion using millions of photographs","volume":"26","author":"Hays","year":"2007","journal-title":"ACM Trans. Graph."},{"key":"B17","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1016\/j.nicl.2017.08.017","article-title":"Identification of autism spectrum disorder using deep learning and the ABIDE dataset","volume":"17","author":"Heinsfeld","year":"2018","journal-title":"NeuroImage"},{"key":"B18","article-title":"\u201cbeta-vae: learning basic visual concepts with a constrained variational framework,\u201d","volume-title":"International Conference on Learning Representations","author":"Higgins","year":"2017"},{"key":"B19","doi-asserted-by":"publisher","first-page":"847","DOI":"10.1016\/j.cviu.2010.04.007","article-title":"Image restoration based on the fast marching method and block based sampling","volume":"114","author":"Huan","year":"2010","journal-title":"Comput. Vis. Image Underst."},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2002.02609","article-title":"Image fine-grained inpainting","author":"Hui","year":"2020","journal-title":"arXiv [preprint]. arXiv:2002.02609"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3072959.3073659","article-title":"Globally and locally consistent image completion","volume":"36","author":"Iizuka","year":"2017","journal-title":"ACM Trans. Graph."},{"key":"B22","doi-asserted-by":"publisher","first-page":"685","DOI":"10.1002\/jmri.21049","article-title":"The Alzheimer's disease neuroimaging initiative (ADNI): MRI methods","volume":"27","author":"Jack Jr","year":"2008","journal-title":"J. Magn. Reson. Imaging"},{"key":"B23","doi-asserted-by":"publisher","first-page":"473","DOI":"10.1109\/42.97598","article-title":"Selection of a convolution function for Fourier inversion using gridding (computerised tomography application)","volume":"10","author":"Jackson","year":"1991","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B24","doi-asserted-by":"publisher","first-page":"14938","DOI":"10.5555\/3540261.3541406","article-title":"Robust compressed sensing MRI with deep generative priors","volume":"34","author":"Jalal","year":"","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2106.11438","article-title":"Instance-optimal compressed sensing via posterior sampling","author":"Jalal","year":"","journal-title":"arXiv [preprint] arXiv:2106.11438"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1807.03039","article-title":"Glow: Generative flow with invertible 1x1 convolutions","author":"Kingma","year":"2018","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B27","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2017.19","article-title":"\u201cPhoto-realistic single image super-resolution using a generative adversarial network,\u201d","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Ledig","year":"2017"},{"key":"B28","doi-asserted-by":"publisher","first-page":"103777","DOI":"10.1016\/j.jvcir.2023.103777","article-title":"Semantic prior-driven fused contextual transformation network for image inpainting","volume":"91","author":"Li","year":"2023","journal-title":"J. Vis. Commun. Image Represent."},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01252-6_6","article-title":"\u201cImage inpainting for irregular holes using partial convolutions,\u201d","author":"Liu","year":"2018","journal-title":"Proceedings of the European Conference on Computer Vision (ECCV)"},{"key":"B30","doi-asserted-by":"crossref","first-page":"725","DOI":"10.1007\/978-3-030-58536-5_43","article-title":"\u201cRethinking image inpainting via a mutual encoder-decoder with feature equalizations,\u201d","volume-title":"Computer Vision\u2013ECCV 2020: 16th European Conference, Glasgow, UK, August 23\u201328, 2020, Proceedings, Part II 16","author":"Liu","year":"2020"},{"key":"B31","first-page":"11461","article-title":"\u201cRepaint: inpainting using denoising diffusion probabilistic models,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Lugmayr","year":"2022"},{"key":"B32","doi-asserted-by":"crossref","first-page":"715","DOI":"10.1007\/978-3-030-58558-7_42","article-title":"\u201cSrflow: learning the super-resolution space with normalizing flow,\u201d","volume-title":"Computer Vision\u2013ECCV 2020: 16th European Conference, Glasgow, UK, August 23\u201328, 2020, Proceedings, Part V 16","author":"Lugmayr","year":"2020"},{"key":"B33","doi-asserted-by":"publisher","first-page":"2677","DOI":"10.1162\/jocn.2009.21407","article-title":"Open access series of imaging studies: longitudinal MRI data in nondemented and demented older adults","volume":"22","author":"Marcus","year":"2010","journal-title":"J. Cogn. Neurosci."},{"key":"B34","doi-asserted-by":"publisher","first-page":"1460","DOI":"10.1002\/acn3.644","article-title":"The Parkinson's progression markers initiative (PPMI)\u2013establishing a PD biomarker cohort","volume":"5","author":"Marek","year":"2018","journal-title":"Ann. Clin. Transl. Neurol."},{"key":"B35","article-title":"\u201cSdedit: guided image synthesis and editing with stochastic differential equations,\u201d","volume-title":"International Conference on Learning Representations","author":"Meng","year":"2021"},{"key":"B36","first-page":"2437","article-title":"\u201cPulse: self-supervised photo upsampling via latent space exploration of generative models,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Menon","year":"2020"},{"key":"B37","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1145\/3503250","article-title":"Nerf: representing scenes as neural radiance fields for view synthesis","volume":"65","author":"Mildenhall","year":"2021","journal-title":"Commun. ACM"},{"key":"B38","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/CSICC49403.2020.9050116","article-title":"\u201cImage inpainting by adaptive fusion of variable spline interpolations,\u201d","volume-title":"2020 25th International Computer Conference, Computer Society of Iran (CSICC)","author":"Nabizadeh","year":"2020"},{"key":"B39","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1901.00212","article-title":"Edgeconnect: generative image inpainting with adversarial edge learning","author":"Nazeri","year":"2019","journal-title":"arXiv [preprint]. arXiv:1901.00212"},{"key":"B40","first-page":"716","article-title":"\u201cAIM 2020 challenge on image extreme inpainting,\u201d","volume-title":"Computer Vision\u2013ECCV 2020 Workshops: Glasgow, UK, August 23\u201328, 2020, Proceedings, Part III 16","author":"Ntavelis","year":""},{"key":"B41","first-page":"394","article-title":"\u201cSesame: semantic editing of scenes by adding, manipulating or erasing objects,\u201d","volume-title":"Computer Vision\u2013ECCV 2020: 16th European Conference, Glasgow, UK, August 23\u201328, 2020, Proceedings, Part XXII 16","author":"Ntavelis","year":""},{"key":"B42","doi-asserted-by":"publisher","first-page":"7474","DOI":"10.1109\/TPAMI.2021.3115428","article-title":"Exploiting deep generative prior for versatile image restoration and manipulation","volume":"44","author":"Pan","year":"2021","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"B43","first-page":"2536","article-title":"\u201cContext encoders: feature learning by inpainting,\u201d","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Pathak","year":"2016"},{"key":"B44","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01042","article-title":"\u201cHigh-resolution image synthesis with latent diffusion models,\u201d","author":"Rombach","year":"2022","journal-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition"},{"key":"B45","doi-asserted-by":"publisher","first-page":"23016","DOI":"10.1117\/1.JEI.30.2.023016","article-title":"Image inpainting using frequency-domain priors","volume":"30","author":"Roy","year":"2021","journal-title":"J. Electron. Imaging"},{"key":"B46","article-title":"\u201cDeep unsupervised learning using nonequilibrium thermodynamics,\u201d","volume-title":"International Conference on Machine Learning","author":"Sohl-Dickstein","year":"2015"},{"key":"B47","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2011.13456","article-title":"Score-based generative modeling through stochastic differential equations","author":"Song","year":"2020","journal-title":"arXiv [preprint]. arXiv:2011.13456"},{"key":"B48","first-page":"2149","article-title":"\u201cResolution-robust large mask inpainting with Fourier convolutions,\u201d","volume-title":"Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision","author":"Suvorov","year":"2022"},{"key":"B49","doi-asserted-by":"crossref","DOI":"10.1109\/CVPRW.2017.149","article-title":"\u201cNtire 2017 challenge on single image super-resolution: methods and results,\u201d","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops","author":"Timofte","year":"2017"},{"key":"B50","first-page":"9446","article-title":"\u201cDeep image prior,\u201d","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Ulyanov","year":"2018"},{"key":"B51","article-title":"Conditional image generation with pixelcnn decoders","author":"Van den Oord","year":"","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B52","first-page":"1747","article-title":"\u201cPixel recurrent neural networks,\u201d","volume-title":"International Conference on Machine Learning","author":"Van Den Oord","year":""},{"key":"B53","doi-asserted-by":"publisher","first-page":"1784","DOI":"10.1109\/TIP.2020.3048629","article-title":"Dynamic selection network for image inpainting","volume":"30","author":"Wang","year":"2021","journal-title":"IEEE Trans. Image Process."},{"key":"B54","doi-asserted-by":"crossref","DOI":"10.1109\/ICCV.2015.164","article-title":"\u201cHolistically-nested edge detection,\u201d","volume-title":"Proceedings of the IEEE International Conference on Computer Vision","author":"Xie","year":"2015"},{"key":"B55","doi-asserted-by":"publisher","first-page":"1308","DOI":"10.1109\/TCSVT.2020.3001267","article-title":"E2I: generative inpainting from edge to image","volume":"31","author":"Xu","year":"2020","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"B56","first-page":"18381","article-title":"\u201cPaint by example: exemplar-based image editing with diffusion models,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Yang","year":"2023"},{"key":"B57","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00457","article-title":"\u201cFree-form image inpainting with gated convolution,\u201d","author":"Yu","year":"2019","journal-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision"},{"key":"B58","first-page":"5505","article-title":"\u201cGenerative image inpainting with contextual attention,\u201d","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Yu","year":"2018"},{"key":"B59","doi-asserted-by":"publisher","first-page":"3266","DOI":"10.1109\/TVCG.2022.3156949","article-title":"Aggregated contextual transformations for high-resolution image inpainting","volume":"29","author":"Zeng","year":"2022","journal-title":"IEEE Trans. Vis. Comput. Graph."},{"key":"B60","first-page":"1","article-title":"\u201cHigh-resolution image inpainting with iterative confidence feedback and guided upsampling,\u201d","volume-title":"Computer Vision\u2013ECCV 2020: 16th European Conference, Glasgow, UK, August 23\u201328, 2020, Proceedings, Part XIX 16","author":"Zeng","year":"2020"},{"key":"B61","doi-asserted-by":"publisher","first-page":"169","DOI":"10.1007\/978-981-32-9244-4_24","article-title":"A brief review of image restoration techniques based on generative adversarial models","volume":"13","author":"Zhang","year":"2020","journal-title":"Advanced Multimedia and Ubiquitous Engineering: MUE\/FutureTech"},{"key":"B62","doi-asserted-by":"crossref","first-page":"1939","DOI":"10.1145\/3240508.3240625","article-title":"\u201cSemantic image inpainting with progressive generative networks,\u201d","volume-title":"Proceedings of the 26th ACM International Conference on Multimedia","author":"Zhang","year":"2018"},{"key":"B63","doi-asserted-by":"publisher","first-page":"117486","DOI":"10.1109\/ACCESS.2021.3106124","article-title":"Automatic object removal with obstructed fa\u00e7ades completion using semantic segmentation and generative adversarial inpainting","volume":"9","author":"Zhang","year":"2021","journal-title":"IEEE Access"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1614608\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,11]],"date-time":"2025-07-11T08:22:48Z","timestamp":1752222168000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1614608\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,11]]},"references-count":63,"alternative-id":["10.3389\/frai.2025.1614608"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1614608","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,11]]},"article-number":"1614608"}}