几何神经距离场用于学习人体运动先验
Geometric Neural Distance Fields for Learning Human Motion Priors
将人体运动先验建模为关节旋转、角速度、角加速度乘积空间上的神经距离场零水平集，为从部分2D/3D观测恢复3D人体运动提供了可复用的几何约束与自适应投影算法。
原文依据 explicitly models the human motion in the zero level set of a collection of neural distance fields (NDFs) corresponding to pose, transition (velocity), and acceleration dynamics
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出神经黎曼运动场（Neural Riemannian Motion Fields），一种新颖的3D生成式人体运动先验，能够实现鲁棒、时序一致且物理合理的3D运动恢复。与现有的基于变分自编码器或扩散模型的方法不同，我们的高阶运动先验在一组神经距离场（NDF）的零水平集上显式建模人体运动，这些距离场分别对应姿态、过渡（速度）和加速度动力学。我们的框架是严谨的，因为我们的NDF构建在关节旋转、其角速度和角加速度的乘积空间上，尊重了底层关节的几何结构。我们进一步引入：(i) 一种新颖的自适应步长混合算法，用于投影到合理运动集合上；(ii) 一种新颖的几何积分器，用于在测试时优化和生成过程中“展开”真实的运动轨迹。我们的实验显示出显著且一致的增益：在AMASS数据集上训练后，神经黎曼运动场在多种输入模态上以及从去噪到运动中间帧生成和拟合部分2D/3D观测等多样化任务上均表现出卓越的泛化能力。
英文原文摘要 1163 字符
We introduce Neural Riemannian Motion Fields (\name), a novel 3D generative human motion prior that enables robust, temporally consistent, and physically plausible 3D motion recovery. Unlike existing VAE or diffusion-based methods, our higher-order motion prior explicitly models the human motion in the zero level set of a collection of neural distance fields (NDFs) corresponding to pose, transition (velocity), and acceleration dynamics. Our framework is rigorous in the sense that our NDFs are constructed on the product space of joint rotations, their angular velocities, and angular accelerations, respecting the geometry of the underlying articulations. We further introduce: (i) a novel adaptive-step hybrid algorithm for projecting onto the set of plausible motions, and (ii) a novel geometric integrator to "roll out" realistic motion trajectories during test-time-optimization and generation. Our experiments show significant and consistent gains: trained on the AMASS dataset, \name remarkably generalizes across multiple input modalities and to diverse tasks ranging from denoising to motion in-betweening and fitting to partial 2D / 3D observations.