Manga2Animate: An Agentic Framework for Faithful Manga Animation with Script and Storyboard Planning
Sheng-Wei Zhang, Hsi-An Chen, Yan-Ting Liou, Chih-Hsuan Chen, Shih-Chen Tseng, Yu-Lun Liu
National Yang Ming Chiao Tung University
Published October 19, 2026 · on the feed October 20, 2026 · ▲ 33 · arXiv 2610.26851
The abstract
Adapting existing manga into coherent animation that stays faithful to the source is a valuable yet challenging task. Existing methods either access the manga mainly through text, losing its characters and art style, or feed manga pages directly into video generation models, which misinterpret the manga's structure and produce incorrect storylines. In this paper, we propose Manga2Animate, a multi-agent framework that grounds every stage of generation in the source manga. We decompose the task into two intermediate representations, an animation script and a visual storyboard. Each representation is produced by a creator agent and reviewed against the source panels by a paired reviewer agent. We tailor representation-specific selection and dual-level feedback to progressively refine both before synthesizing the final animation. Extensive experiments and a user study demonstrate that Manga2Animate produces animations more faithful to the source manga and more consistent across shots than state-of-the-art approaches. Project page: https://annchenn.github.io/Manga2Animate/
The Brief
1:54A short podcast summary of the paper. Written and voiced by AI.
Transcript
Manga2Animate. An Agentic Framework for Faithful Manga Animation with Script and Storyboard Planning. By Sheng-Wei Zhang and five others, from National Yang Ming Chiao Tung University. Published on the nineteenth of October. Its code repository is worked out: about three hundred source files, with a readme.
Adapting existing manga into animation that is coherent, and stays faithful to its source, is a valuable task and a challenging one. Existing methods fall short in one of two ways. Some access the manga mainly through text, and lose its characters and its art style along the way. Others feed manga pages directly into video generation models, which misinterpret the manga's structure and produce incorrect storylines. One route keeps the story and drops the pictures. The other keeps the pictures and garbles the story.
Manga2Animate is a multi-agent framework that grounds every stage of generation in the source manga. Rather than going from page to video in one step, it decomposes the task into two intermediate representations: an animation script and a visual storyboard.
Each representation is produced by a creator agent, and reviewed against the source panels by a paired reviewer agent. Selection is tailored to each representation, and feedback works at two levels. Together they progressively refine both the script and the storyboard, and only after that is the final animation synthesized.
Extensive experiments and a user study show that Manga2Animate produces animations more faithful to the source manga, and more consistent across shots, than state-of-the-art approaches.
Each failure named at the outset meets a part of the design. Text-only methods lost the visuals, and here every review is made against the source panels themselves. Video models misread the page structure, and here a script and a storyboard set out that structure before any video is generated. The pairing of creator and reviewer is what keeps each intermediate step tied to the manga it came from.
More with an account
Listen to every day’s papers
Daily playlists, deep dives, quests for the papers, and quizzes on their key terms.
Join free