Search papers, labs, and topics across Lattice.
MoZoo is introduced as a generative dynamics solver that synthesizes high-fidelity animal videos from coarse meshes using diffusion models, bypassing traditional refinement processes. The method incorporates Role-Aware RoPE (RAR-RoPE) for motion synchronization and Asymmetric Decoupled Attention to prevent feature interference and improve efficiency. To overcome data scarcity, the authors also create MoZoo-Data, a synthetic-to-real pipeline for generating paired sequences, and MoZooBench, a benchmark with 120 mesh-video pairs, demonstrating superior fur simulation across various animal skeletons.
Synthesizing realistic animal fur and muscle dynamics is now possible directly from coarse meshes, thanks to a novel diffusion-based approach that leapfrogs traditional, computationally expensive refinement methods.
The creation of cinematic-quality animal effects necessitates the precise modeling of muscle and fur dynamics, a process that remains both labor-intensive and computationally expensive within traditional production workflows. While generative diffusion models have shown promise in diverse artistic workflows, their capacity for high-fidelity animal simulation remains largely unexploited. We present MoZoo, a generative dynamics solver that bypasses conventional refinement to synthesize high-fidelity animal videos from coarse meshes under multimodal guidance. We propose Role-Aware RoPE (RAR-RoPE) which employs role-based index remapping to synchronize motion alignment while decoupling reference information via fixed temporal offsets. Complementing this, Asymmetric Decoupled Attention partitions the latent sequence to enforce a unidirectional information flow, effectively preventing feature interference and improving computational efficiency. To address the scarcity of high-quality training data, we introduce MoZoo-Data, a synthetic-to-real pipeline that leverages a rendering engine and an inverse mapping approach to construct a large-scale dataset of paired sequences. Furthermore, we establish MoZooBench, a comprehensive benchmark with 120 mesh-video pairs. Experimental results demonstrate that MoZoo achieves high-fidelity fur simulation across diverse animal skeletons and layouts, preserving superior temporal and structural consistency.
