New AI framework CreatiPoster generates editable multi-layer graphic designs

Researchers have introduced CreatiPoster, a novel framework designed to generate editable, multi-layer graphic designs from natural language instructions or provided assets. Unlike existing tools that struggle with user asset integration and editability, CreatiPoster utilizes a protocol model and an RGBA large multimodal model to create a JSON specification for each design layer. This specification includes layout, content, style, and background details, enabling coherent background synthesis. The framework aims to advance AI-assisted graphic design by supporting applications like canvas editing, text overlay, and responsive resizing, and the researchers have released a corpus of 100,000 multi-layer designs to foster further research. AI

IMPACT This framework could democratize graphic design by enabling users to create editable, multi-layer compositions with AI assistance.

RANK_REASON The cluster describes a new research paper introducing a novel AI framework for graphic design generation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• Canva Magic Design
• CreatiPoster
• Dexiang Hong
• JSON

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →