AnimeAgent: Is the Multi-Agent via Image-to-Video models a Good Disney Storytelling Artist?

TL;DR

AnimeAgent introduces an innovative Image-to-Video multi-agent framework for custom storyboard generation, overcoming static model limitations and enabling iterative refinement for high-quality, stylized storytelling.

Contribution

It is the first I2V-based multi-agent system for CSG, inspired by Disney's animation workflow, with a new benchmark and a mixed reviewer for reliable iterative improvements.

Findings

Achieves state-of-the-art consistency and stylization.

Improves prompt fidelity in storyboard generation.

Provides a new human-annotated CSG benchmark.

Abstract

Custom Storyboard Generation (CSG) aims to produce high-quality, multi-character consistent storytelling. Current approaches based on static diffusion models, whether used in a one-shot manner or within multi-agent frameworks, face three key limitations: (1) Static models lack dynamic expressiveness and often resort to "copy-paste" pattern. (2) One-shot inference cannot iteratively correct missing attributes or poor prompt adherence. (3) Multi-agents rely on non-robust evaluators, ill-suited for assessing stylized, non-realistic animation. To address these, we propose AnimeAgent, the first Image-to-Video (I2V)-based multi-agent framework for CSG. Inspired by Disney's "Combination of Straight Ahead and Pose to Pose" workflow, AnimeAgent leverages I2V's implicit motion prior to enhance consistency and expressiveness, while a mixed subjective-objective reviewer enables reliable iterative refinement. We also collect a human-annotated CSG benchmark with ground-truth. Experiments show AnimeAgent achieves SOTA performance in consistency, prompt fidelity, and stylization.