Transformer-based advertisement image layout generation and object fidelity optimization
PLOS One · 25.06.2026 · 21 (6) · e0352418
DOI: 10.1371/journal.pone.0352418 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 25.06.2026. Verfasst von Dandan Xu und 2 weiteren Autor:innen. Beteiligte Einrichtung: College of Plastic Arts, Daegu University, Gyeongsan-si, Gyeongsangbuk-do, Republic of Korea. Hauptthemen laut MEDLINE: Advertising, Image Processing, Computer-Assisted. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The layout of the image and the positional distribution of the objects directly affect the audience’s visual focus and the effect of the message conveyed. By setting up a specific layout structure, the viewer’s attention can be more focused on the product launched or the moral of the advertisement. Existing work related to position-controllable text-to-image generation 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–layout 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.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- College of Plastic Arts, Daegu University, Gyeongsan-si, Gyeongsangbuk-do, Republic of Korea.
- College of Computer Science and Mathematics, Fujian University of Technology, Fuzhou, Fujian, China.
- Hanyang University ERICA Campus, Ansan-si, Gyeonggi-do, South Korea.
- Xing'an Polytechnic University, Dulin West Street, Horqin Right Front Banner, Xing'an League, Inner Mongolia, P.R. China.
ORCID-Profile: Linlu Cai
Zitieren
APA
Xu, D., Hu, K., & Cai, L. (2026). Transformer-based advertisement image layout generation and object fidelity optimization. PLOS One, 21(6), e0352418. https://doi.org/10.1371/journal.pone.0352418
Vancouver
Xu D, Hu K, Cai L. Transformer-based advertisement image layout generation and object fidelity optimization. PLOS One. 2026;21(6):e0352418. doi:10.1371/journal.pone.0352418
BibTeX
@article{Xu2026,
author = {Xu, Dandan and Hu, Kun and Cai, Linlu},
title = {Transformer-based advertisement image layout generation and object fidelity optimization},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {6},
pages = {e0352418},
doi = {10.1371/journal.pone.0352418}
}
EUVIMED