MikuDance: Animating Character Art with Mixed Motion Dynamics

TL;DR

MikuDance introduces a diffusion-based pipeline with mixed motion techniques to animate stylized characters, effectively handling high-dynamic motions and reference-guidance issues for high-quality, flexible character animation.

Contribution

It presents novel mixed motion modeling and control diffusion techniques, including scene motion tracking and motion-adaptive normalization, to improve character art animation quality and flexibility.

Findings

Effective handling of high-dynamic motion in character animation

Generalizes well across various character styles and motions

Produces high-quality, realistic animations with complex dynamics

Abstract

We propose MikuDance, a diffusion-based pipeline incorporating mixed motion dynamics to animate stylized character art. MikuDance consists of two key techniques: Mixed Motion Modeling and Mixed-Control Diffusion, to address the challenges of high-dynamic motion and reference-guidance misalignment in character art animation. Specifically, a Scene Motion Tracking strategy is presented to explicitly model the dynamic camera in pixel-wise space, enabling unified character-scene motion modeling. Building on this, the Mixed-Control Diffusion implicitly aligns the scale and body shape of diverse characters with motion guidance, allowing flexible control of local character motion. Subsequently, a Motion-Adaptive Normalization module is incorporated to effectively inject global scene motion, paving the way for comprehensive character art animation. Through extensive experiments, we demonstrate the effectiveness and generalizability of MikuDance across various character art and motion guidance, consistently producing high-quality animations with remarkable motion dynamics.