Automatic Text Box Placement for Supporting Typographic Design

2025/10/09 by Jun Muraoka, Muraoka, Jun, Daichi Haraguchi +9
Computer Science · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Handwritten Text Recognition Techniques #Interactive and Immersive Displays #Multimodal Machine Learning Applications

paper · pdf · doi:10.48550/arxiv.2510.07665

openalex publication_date 2025/10/09 · openalex created_date 2025/10/11 · openalex updated_date 2026/07/28

Abstract

In layout design for advertisements and web pages, balancing visual appeal and communication efficiency is crucial. This study examines automated text box placement in incomplete layouts, comparing a standard Transformer-based method, a small Vision and Language Model (Phi3.5-vision), a large pretrained VLM (Gemini), and an extended Transformer that processes multiple images. Evaluations on the Crello dataset show the standard Transformer-based models generally outperform VLM-based approaches, particularly when incorporating richer appearance information. However, all methods face challenges with very small text or densely populated layouts. These findings highlight the benefits of task-specific architectures and suggest avenues for further improvement in automated layout design.

Citations

• PosterLLaVa: Constructing a Unified Multi-modal Layout Generator with LLM
• Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
• PosterLlama: Bridging Design Ability of Langauge Model to Contents-Aware Layout Generation
• Retrieval-Augmented Layout Transformer for Content-Aware Layout Generation
• LayoutPrompter: Awaken the Design Ability of Large Language Models
• Relation-Aware Diffusion Model for Controllable Poster Layout Generation
• LayoutGPT: Compositional Visual Planning and Generation with Large Language Models
• Towards Flexible Multi-modal Document Models
• GPT-4 Technical Report
• LayoutDM: Discrete Diffusion Model for Controllable Layout Generation
• CanvasVAE: Learning to Generate Vector Graphic Documents
• LoRA: Low-Rank Adaptation of Large Language Models
• Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression
• Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression
• LayoutVAE: Stochastic Scene Layout Generation From a Label Set
• LayoutGAN: Generating Graphic Layouts with Wireframe Discriminators
• COLE: A Hierarchical Generation Framework for Multi-Layered and Editable Graphic Design

Related

• Content-Aware Ad Banner Layout Generation with Two-Stage Chain-of-Thought in Vision Language Models
• LayoutDiT: Exploring Content-Graphic Balance in Layout Generation with Diffusion Transformer
• TextLap: Customizing Language Models for Text-to-Layout Planning
• VASCAR: Content-Aware Layout Generation via Visual-Aware Self-Correction
• The Layout Generation Algorithm of Graphic Design Based on Transformer-CVAE