DiffSensei: Bridging Multi-Modal LLMs and Diffusion Models for Customized Manga Generation

TL;DR

DiffSensei is a novel framework that combines diffusion models and multimodal large language models to enable highly customizable manga generation with dynamic multi-character control from textual descriptions.

Contribution

The paper introduces DiffSensei, a new method integrating diffusion models with MLLMs for precise, flexible manga creation, and provides MangaZero, a large-scale dataset for this task.

Findings

DiffSensei outperforms existing models in manga generation quality.

It enables flexible character customization based on text cues.

The approach effectively manages multi-character interactions and expressions.

Abstract

Story visualization, the task of creating visual narratives from textual descriptions, has seen progress with text-to-image generation models. However, these models often lack effective control over character appearances and interactions, particularly in multi-character scenes. To address these limitations, we propose a new task: \textbf{customized manga generation} and introduce \textbf{DiffSensei}, an innovative framework specifically designed for generating manga with dynamic multi-character control. DiffSensei integrates a diffusion-based image generator with a multimodal large language model (MLLM) that acts as a text-compatible identity adapter. Our approach employs masked cross-attention to seamlessly incorporate character features, enabling precise layout control without direct pixel transfer. Additionally, the MLLM-based adapter adjusts character features to align with panel-specific text cues, allowing flexible adjustments in character expressions, poses, and actions. We also introduce \textbf{MangaZero}, a large-scale dataset tailored to this task, containing 43,264 manga pages and 427,147 annotated panels, supporting the visualization of varied character interactions and movements across sequential frames. Extensive experiments demonstrate that DiffSensei outperforms existing models, marking a significant advancement in manga generation by enabling text-adaptable character customization. The project page is https://jianzongwu.github.io/projects/diffsensei/.