AnyMoLe: Any Character Motion In-betweening Leveraging Video Diffusion Models
This addresses the data dependency problem in motion in-betweening for animation and gaming applications, representing a novel method rather than an incremental improvement.
The paper tackles the limitation of requiring character-specific datasets for motion in-betweening by introducing AnyMoLe, which leverages video diffusion models to generate motion for arbitrary characters without external data, achieving smooth and realistic transitions.
Despite recent advancements in learning-based motion in-betweening, a key limitation has been overlooked: the requirement for character-specific datasets. In this work, we introduce AnyMoLe, a novel method that addresses this limitation by leveraging video diffusion models to generate motion in-between frames for arbitrary characters without external data. Our approach employs a two-stage frame generation process to enhance contextual understanding. Furthermore, to bridge the domain gap between real-world and rendered character animations, we introduce ICAdapt, a fine-tuning technique for video diffusion models. Additionally, we propose a ``motion-video mimicking'' optimization technique, enabling seamless motion generation for characters with arbitrary joint structures using 2D and 3D-aware features. AnyMoLe significantly reduces data dependency while generating smooth and realistic transitions, making it applicable to a wide range of motion in-betweening tasks.