CreatiParser: Generative Image Parsing of Raster Graphic Designs into Editable Layers

TL;DR

CreatiParser introduces a hybrid generative approach for converting raster graphic designs into editable layers, improving editing flexibility and accuracy over existing multi-stage methods.

Contribution

The paper presents a novel raster-to-layer parsing framework combining vision-language models and diffusion architectures, with a new reward mechanism for aligning with human preferences.

Findings

Achieves 23.7% average improvement over previous methods.

Effectively decomposes images into editable text, background, and sticker layers.

Demonstrates superior performance on Parser-40K and Crello datasets.

Abstract

Graphic design images consist of multiple editable layers, such as text, background, and decorative elements, while most generative models produce rasterized outputs without explicit layer structures, limiting downstream editing. Existing graphic design parsing methods typically rely on multi-stage pipelines combining layout prediction, matting, and inpainting, which suffer from error accumulation and limited controllability. We propose a hybrid generative framework for raster-to-layer graphic design parsing that decomposes a design image into editable text, background, and sticker layers. Text regions are parsed using a vision-language model into a text rendering protocol, enabling faithful reconstruction and flexible re-editing, while background and sticker layers are generated using a multi-branch diffusion architecture with RGBA support. We further introduce ParserReward and integrate it with Group Relative Policy Optimization to align generation quality with human design preferences. Extensive experiments on two challenging datasets, \emph{i.e.,} the Parser-40K and Crello datasets, demonstrate superior performance over existing methods, \emph{eg.,} achieving an overall average improvement of 23.7\% across all metrics.