用视频扩散模型蒸馏出高保真角色动画，兼顾运动质量和形状一致性。
Articulated Kinematics Distillation from Video Diffusion Models
- 基于骨骼控制降低自由度，结合扩散模型生成动作
- 在文本到4D生成任务中实现更优的3D一致性和运动质量
- 适合需要物理合理交互的动画生成场景
我们提出一种名为关节运动知识蒸馏（AKD）的框架，通过融合基于骨架的动画与现代生成模型的优势，生成高保真角色动画。AKD采用骨架表示法处理带绑定的3D资产，大幅降低自由度（DoFs），聚焦于关节级控制，实现高效且一致的运动合成。利用预训练视频扩散模型的分数蒸馏采样（SDS），AKD在保持结构完整性的同时，蒸馏出复杂的关节式运动，克服了4D神经变形场在保持形状一致性方面的挑战。该方法天然兼容物理仿真，确保物理合理的交互。实验表明，相比现有方法，AKD在文本到4D生成任务中实现了更优的3D一致性和运动质量。
原文摘要 · Abstract (English)
We present Articulated Kinematics Distillation (AKD), a framework for generating high-fidelity character animations by merging the strengths of skeleton-based animation and modern generative models. AKD uses a skeleton-based representation for rigged 3D assets, drastically reducing the Degrees of Freedom (DoFs) by focusing on joint-level control, which allows for efficient, consistent motion synthesis. Through Score Distillation Sampling (SDS) with pre-trained video diffusion models, AKD distills complex, articulated motions while maintaining structural integrity, overcoming challenges faced by 4D neural deformation fields in preserving shape consistency. This approach is naturally compatible with physics-based simulation, ensuring physically plausible interactions. Experiments show that AKD achieves superior 3D consistency and motion quality compared with existing works on text-to-4D generation. Project page: https://research.nvidia.com/labs/dir/akd/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。