CreatiPoster: Towards Editable and Controllable Multi-Layer Graphic Design Generation

2025-06-12

AI Summary

To address the high skill barrier, poor editability, and low-quality multi-layer generation in graphic design, this paper introduces the first end-to-end framework for editable multi-layer poster generation. Methodologically: (1) we propose the first RGBA-aware large vision-language model that directly outputs structured JSON design specifications—including precise layout, layer hierarchy, styling attributes, and per-pixel RGBA channel information; (2) we introduce a foreground-background decoupled generation paradigm, integrating a conditional background synthesis model to enhance visual consistency; (3) we construct the first open-source, copyright-free, 100K-scale multi-layer design benchmark and corpus. Experiments demonstrate that our approach comprehensively outperforms state-of-the-art commercial systems (e.g., Canva Magic Design) in generation quality, editability, and controllability—enabling canvas-level editing, multilingual instruction following, responsive scaling, and dynamic poster generation.

Technology Category

• Computer Vision: Large Vision Models
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Graphic design plays a crucial role in both commercial and personal contexts, yet creating high-quality, editable, and aesthetically pleasing graphic compositions remains a time-consuming and skill-intensive task, especially for beginners. Current AI tools automate parts of the workflow, but struggle to accurately incorporate user-supplied assets, maintain editability, and achieve professional visual appeal. Commercial systems, like Canva Magic Design, rely on vast template libraries, which are impractical for replicate. In this paper, we introduce CreatiPoster, a framework that generates editable, multi-layer compositions from optional natural-language instructions or assets. A protocol model, an RGBA large multimodal model, first produces a JSON specification detailing every layer (text or asset) with precise layout, hierarchy, content and style, plus a concise background prompt. A conditional background model then synthesizes a coherent background conditioned on this rendered foreground layers. We construct a benchmark with automated metrics for graphic-design generation and show that CreatiPoster surpasses leading open-source approaches and proprietary commercial systems. To catalyze further research, we release a copyright-free corpus of 100,000 multi-layer designs. CreatiPoster supports diverse applications such as canvas editing, text overlay, responsive resizing, multilingual adaptation, and animated posters, advancing the democratization of AI-assisted graphic design. Project homepage: https://github.com/graphic-design-ai/creatiposter

Problem

• Automating high-quality editable multi-layer graphic design generation
• Enhancing user control over design assets and editability
• Overcoming limitations of template-based commercial design systems

Innovation

• Generates editable multi-layer designs from instructions
• Uses RGBA model for precise layer specifications
• Conditional background model ensures visual coherence