{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,25]],"date-time":"2026-06-25T17:50:59Z","timestamp":1782409859812,"version":"3.54.5"},"reference-count":44,"publisher":"Public Library of Science (PLoS)","issue":"6","license":[{"start":{"date-parts":[[2026,6,25]],"date-time":"2026-06-25T00:00:00Z","timestamp":1782345600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>The layout of the image and the positional distribution of the objects directly affect the audience\u2019s visual focus and the effect of the message conveyed. By setting up a specific layout structure, the viewer\u2019s attention can be more focused on the product launched or the moral of the advertisement. Existing work related to position-controllable text-to-image\u00a0generation has made great progress in generating results on simple images. However, when generating images of complex scenes, the quality is often poor. This can result in the model failing to accurately convey the message of the advertisement when used to generate advertisements. To address these limitations, we propose LAYOBJ-GAN, a novel two-stage framework for layout-controllable advertisement image synthesis. Unlike prior work, our method explicitly models background layouts jointly with object layouts during the text-to-layout generation stage, enabling comprehensive spatial planning of complex scenes. Technically, we introduce a Transformer-based sequence-to-sequence layout generator that learns long-range dependencies between textual descriptions and both background and object regions, which has not been explored in previous advertisement-oriented text-to-image frameworks. In the layout-to-image stage, we further propose a fine-grained text\u2013layout interaction normalization module (TL-Norm) that enables object knowledge transfer from a pre-trained category-to-image model, allowing object appearance to be adaptively modulated by textual context and layout constraints. Extensive experiments on MS-COCO and a high-definition advertisement dataset (AsHQ-10K) demonstrate that LAYOBJ-GAN significantly outperforms seven state-of-the-art methods in image quality, layout controllability, and semantic object accuracy. These results confirm the effectiveness of explicitly modeling background layouts and transferring object-level generative knowledge for complex advertisement image synthesis.<\/jats:p>","DOI":"10.1371\/journal.pone.0352418","type":"journal-article","created":{"date-parts":[[2026,6,25]],"date-time":"2026-06-25T17:41:49Z","timestamp":1782409309000},"page":"e0352418","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Transformer-based advertisement image layout generation and object fidelity optimization"],"prefix":"10.1371","volume":"21","author":[{"given":"Dandan","family":"Xu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kun","family":"Hu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-7942-7965","authenticated-orcid":true,"given":"Linlu","family":"Cai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,6,25]]},"reference":[{"key":"pone.0352418.ref001","doi-asserted-by":"crossref","first-page":"1505","DOI":"10.1109\/CVPR.2019.00160","article-title":"MirrorGAN: learning text-to-image generation by redescription","author":"T Qiao","year":"2019","journal-title":"2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"pone.0352418.ref002","first-page":"1886","article-title":"Ad Lingua: text classification improves symbolism prediction in image advertisements.","author":"A Savchenko","year":"2020","journal-title":"Proceedings of the 28th International Conference on Computational Linguistics. Barcelona, Spain (Online): International Committee on Computational Linguistics"},{"key":"pone.0352418.ref003","doi-asserted-by":"crossref","first-page":"111013","DOI":"10.1016\/j.patcog.2024.111013","article-title":"Scene Chinese recognition with local and global attention","volume":"158","author":"Z Chen","year":"2025","journal-title":"Pattern Recognit"},{"key":"pone.0352418.ref004","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TITS.2025.3534291","article-title":"Arbitrary shape text detection with discrete cosine transform and CLIP for urban scene perception in ITS","author":"Z Chen","year":"2025","journal-title":"IEEE Trans Intell Transp Syst"},{"key":"pone.0352418.ref005","doi-asserted-by":"crossref","first-page":"22","DOI":"10.1080\/00913367.2021.1909515","article-title":"Preparing for an era of deepfakes and AI-generated ads: a framework for understanding responses to manipulated advertising","volume":"51","author":"C Campbell","year":"2022","journal-title":"J Advert"},{"key":"pone.0352418.ref006","author":"IJ Goodfellow","year":"2014"},{"key":"pone.0352418.ref007","author":"L Zhang","year":"2024"},{"key":"pone.0352418.ref008","doi-asserted-by":"crossref","first-page":"740","DOI":"10.1007\/978-3-319-10602-1_48","article-title":"Microsoft COCO: common objects in context.","volume-title":"Computer vision \u2013 ECCV 2014","author":"T-Y Lin","year":"2014"},{"key":"pone.0352418.ref009","doi-asserted-by":"crossref","first-page":"595","DOI":"10.1109\/CVPR.2015.7298658","article-title":"Building a bird recognition app and large scale dataset with citizen scientists: the fine print in fine-grained dataset collection","author":"G Van Horn","year":"2015","journal-title":"2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"pone.0352418.ref010","doi-asserted-by":"crossref","first-page":"121","DOI":"10.1007\/978-3-030-20656-7_7","article-title":"Layout-aware embedding for quantum annealing processors.","volume-title":"High performance computing","author":"JP Pinilla","year":"2019"},{"issue":"4","key":"pone.0352418.ref011","doi-asserted-by":"crossref","first-page":"100089","DOI":"10.1016\/j.tbench.2023.100089","article-title":"An era of ChatGPT as a significant futuristic support tool: a study on features, abilities, and challenges","volume":"2","author":"A Haleem","year":"2022","journal-title":"BenchCouncil Trans Benchmarks Stand Eval"},{"key":"pone.0352418.ref012","first-page":"1","article-title":"The impact of AI on the advertising process: the Chinese experience","volume":"48","author":"X Qin","year":"2019","journal-title":"J Advert"},{"key":"pone.0352418.ref013","doi-asserted-by":"crossref","first-page":"244","DOI":"10.3390\/digital4010013","article-title":"Decoding the relationship of artificial intelligence, advertising, and generative models","volume":"4","author":"CV Lim","year":"2024","journal-title":"Digital"},{"key":"pone.0352418.ref014","doi-asserted-by":"crossref","first-page":"346","DOI":"10.1109\/ISBDAS64762.2025.11116982","article-title":"Image recognition and extraction of deep features using GoogLeNet","author":"S Kukkadapu","year":"2025","journal-title":"2025 8th International Symposium on Big Data and Applied Statistics (ISBDAS)"},{"key":"pone.0352418.ref015","article-title":"Capsulegan: generative adversarial capsule network","author":"A Jaiswal","year":"2018","journal-title":"Proceedings of the European Conference on Computer Vision (ECCV) Workshops"},{"key":"pone.0352418.ref016","author":"H Yu","year":"2020"},{"key":"pone.0352418.ref017","doi-asserted-by":"crossref","first-page":"18","DOI":"10.1007\/978-3-030-31723-2_2","article-title":"Recurrent deconvolutional generative adversarial networks with application to video generation","author":"H Yu","year":"2019","journal-title":"Pattern Recognition and Computer Vision: Second Chinese Conference, PRCV 2019, Xi\u2019an, China, 2019 Nov 8\u201311, Proceedings, Part II"},{"key":"pone.0352418.ref018","doi-asserted-by":"crossref","first-page":"1526","DOI":"10.1109\/CVPR.2018.00165","article-title":"MoCoGAN: decomposing motion and content for video generation","author":"S Tulyakov","year":"2018","journal-title":"2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition"},{"key":"pone.0352418.ref019","author":"Z Yang","year":"2025"},{"key":"pone.0352418.ref020","author":"AK Akan","year":"2021"},{"key":"pone.0352418.ref021","first-page":"1","article-title":"Identification of DNA motif using likelihood and attention based pooling method in the GRU framework","author":"M Boruah","year":"2021","journal-title":"2021 6th IEEE International Conference on Recent Advances and Innovations in Engineering (ICRAIE)"},{"key":"pone.0352418.ref022","article-title":"Transfer learning-based advanced deep learning architecture for the identification of HIV-1 integration sites using imbalanced dataset.","author":"M Boruah","year":"2023","journal-title":"Intelligent data engineering and analytics. FICTA 2022. smart innovation, systems and technologies, vol 327"},{"issue":"23","key":"pone.0352418.ref023","doi-asserted-by":"crossref","first-page":"17113","DOI":"10.1007\/s00521-023-08585-y","article-title":"CaDenseNet: a novel deep learning approach using capsule network with attention for the identification of HIV-1 integration site","volume":"35","author":"M Boruah","year":"2023","journal-title":"Neural Comput Appl"},{"key":"pone.0352418.ref024","doi-asserted-by":"crossref","first-page":"104990","DOI":"10.1016\/j.imavis.2024.104990","article-title":"MLCapsNet +: a multi-capsule network for the identification of the HIV ISs along important sequence positions","volume":"145","author":"M Boruah","year":"2024","journal-title":"Image Vis Comput"},{"key":"pone.0352418.ref025","article-title":"Attention is all you need.","volume-title":"Advances in neural information processing systems","author":"A Vaswani","year":"2017"},{"key":"pone.0352418.ref026","author":"W Hong","year":"2022"},{"key":"pone.0352418.ref027","author":"J Xu","year":"2024"},{"key":"pone.0352418.ref028","author":"A Polyak","year":"2025"},{"key":"pone.0352418.ref029","author":"U Singer","year":"2022"},{"key":"pone.0352418.ref030","author":"R Villegas","year":"2022"},{"key":"pone.0352418.ref031","author":"J Ho","year":"2022"},{"key":"pone.0352418.ref032","author":"W Yan","year":"2021"},{"key":"pone.0352418.ref033","article-title":"A survey on video diffusion models","volume":"57","author":"Z Xing","year":"2024","journal-title":"ACM Comput Surv"},{"key":"pone.0352418.ref034","author":"O Ronneberger","year":"2015"},{"issue":"12","key":"pone.0352418.ref035","doi-asserted-by":"crossref","first-page":"10076","DOI":"10.1109\/TPAMI.2024.3435571","article-title":"Medical image segmentation review: the success of U-Net","volume":"46","author":"R Azad","year":"2024","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"pone.0352418.ref036","doi-asserted-by":"crossref","first-page":"7827","DOI":"10.1109\/CVPR52733.2024.00748","article-title":"SimDA: simple diffusion adapter for efficient video generation","author":"Z Xing","year":"2024","journal-title":"2024 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"pone.0352418.ref037","article-title":"Make-a-video: text-to-video generation without text-video data","author":"U Singer","year":"2022","journal-title":"ArXiv"},{"key":"pone.0352418.ref038","doi-asserted-by":"crossref","first-page":"22563","DOI":"10.1109\/CVPR52729.2023.02161","article-title":"Align your latents: high-resolution video synthesis with latent diffusion models","author":"A Blattmann","year":"2023","journal-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"pone.0352418.ref039","author":"J Ho","year":"2022"},{"key":"pone.0352418.ref040","author":"OpenAI","year":"2024"},{"key":"pone.0352418.ref041","author":"R Sun","year":"2024"},{"key":"pone.0352418.ref042","author":"J Gu","year":"2023"},{"key":"pone.0352418.ref043","doi-asserted-by":"crossref","unstructured":"Lee T, Kwon S, Kim T. Grid diffusion models for text-to-video generation; 2024. p. 8734\u201343. https:\/\/doi.org\/10.1109\/CVPR52733.2024.00834","DOI":"10.1109\/CVPR52733.2024.00834"},{"key":"pone.0352418.ref044","doi-asserted-by":"crossref","first-page":"10530","DOI":"10.1109\/ICCV.2019.01063","article-title":"Image synthesis from reconfigurable layout and style","author":"W Sun","year":"2019","journal-title":"2019 IEEE\/CVF International Conference on Computer Vision (ICCV)"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0352418","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,25]],"date-time":"2026-06-25T17:41:58Z","timestamp":1782409318000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0352418"}},"subtitle":[],"editor":[{"given":"Zahid","family":"Mehmood","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,6,25]]},"references-count":44,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6,25]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0352418","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,25]]}}}