Decomate: Leveraging Generative Models for Co-Creative SVG Animation

Jihyeon Park, Jiyoon Myung, Seone Shin, JUNGKI SON, Joohyung Han

Track: Short paper

Keywords: co-creation, creativeAI, generativeAI

TL;DR: Decomate is a system that uses multimodal large language models to automatically restructure static SVGs into animation-ready components, allowing designers to create HTML/CSS animations through simple natural language prompts.

Abstract: Designers often face friction when attempting to animate static SVG graphics, particularly when the visual structure does not align with the intended motion granularity. Existing tools typically rely on predefined groupings or require technical fluency, limiting designers’ ability to experiment and iterate independently. We present Decomate, a system that enables intuitive SVG animation through natural language. Decomate first uses a large language model (LLM) to restructure raw SVGs into semantically meaningful, animation-ready components. Designers can then assign motion to each component via text prompts, and the system generates corresponding HTML/CSS animations. Decomate brings generative AI into creative workflows by enabling iterative refinement through natural language based on user intent.

Submission Number: 48