Behave Your Motion: Habit-preserved Cross-category Animal Motion Transfer
This work addresses a critical task for animation and virtual reality applications by enabling cross-category animal motion transfer with preserved habits, representing an incremental advance over existing methods focused on humans.
The paper tackles the problem of transferring motion across animal categories while preserving species-specific habitual behaviors, which is complex for animation and virtual reality, and proposes a novel framework that achieves superior performance validated on a new quadruped dataset.
Animal motion embodies species-specific behavioral habits, making the transfer of motion across categories a critical yet complex task for applications in animation and virtual reality. Existing motion transfer methods, primarily focused on human motion, emphasize skeletal alignment (motion retargeting) or stylistic consistency (motion style transfer), often neglecting the preservation of distinct habitual behaviors in animals. To bridge this gap, we propose a novel habit-preserved motion transfer framework for cross-category animal motion. Built upon a generative framework, our model introduces a habit-preservation module with category-specific habit encoder, allowing it to learn motion priors that capture distinctive habitual characteristics. Furthermore, we integrate a large language model (LLM) to facilitate the motion transfer to previously unobserved species. To evaluate the effectiveness of our approach, we introduce the DeformingThings4D-skl dataset, a quadruped dataset with skeletal bindings, and conduct extensive experiments and quantitative analyses, which validate the superiority of our proposed model.