Retrieval Augmented Comic Image Generation

TL;DR

RaCig is a new system that generates comic-style image sequences with consistent characters and expressive gestures by combining retrieval-based character alignment and regional feature embedding.

Contribution

It introduces a retrieval-based character assignment and regional injection mechanism for coherent and expressive comic image generation.

Findings

Effective generation of comic narratives with consistent characters

Maintains character identity and costume across frames

Produces diverse, vivid character gestures

Abstract

We present RaCig, a novel system for generating comic-style image sequences with consistent characters and expressive gestures. RaCig addresses two key challenges: (1) maintaining character identity and costume consistency across frames, and (2) producing diverse and vivid character gestures. Our approach integrates a retrieval-based character assignment module, which aligns characters in textual prompts with reference images, and a regional character injection mechanism that embeds character features into specified image regions. Experimental results demonstrate that RaCig effectively generates engaging comic narratives with coherent characters and dynamic interactions. The source code will be publicly available to support further research in this area.