We address the problem of fitting a parametric human body model (SMPL) to point cloud data. Optimization-based methods require careful initialization and are prone to becoming trapped in local optima. Learning-based methods address this but do not generalize well when the input pose is far from those seen during training. For rigid point clouds, remarkable generalization has been achieved by leveraging SE(3)-equivariant networks, but these methods do not work on articulated objects. In this work we extend this idea to human bodies and propose ArtEq, a novel part-based SE(3)-equivariant neural architecture for SMPL model estimation from point clouds. Specifically, we learn a part detection network by leveraging local SO(3) invariance, and regress shape and pose using articulated SE(3) shape-invariant and pose-equivariant networks, all trained end-to-end. Our novel equivariant pose regression module leverages the permutation-equivariant property of self-attention layers to preserve rotational equivariance. Experimental results show that ArtEq can generalize to poses not seen during training, outperforming state-of-the-art methods by 74.5%, without requiring an optimization refinement step. Further, compared with competing works, our method is more than three orders of magnitude faster during inference and has 97.3% fewer parameters. The code and model will be available for research purposes at https://arteq.is.tue.mpg.de.

翻译：我们研究了将参数化人体模型（SMPL）拟合到点云数据的问题。基于优化的方法需要精心初始化且易陷入局部最优。基于学习的方法虽能解决此问题，但当输入姿态与训练数据相差较大时泛化能力不足。针对刚性点云，利用SE(3)等变网络已实现卓越泛化，但此类方法不适用于关节物体。本研究将此思想扩展至人体，提出ArtEq——一种基于部件的SE(3)等变神经架构，用于从点云中估计SMPL模型。具体而言，我们利用局部SO(3)不变性学习部件检测网络，并通过关节SE(3)形状不变与姿态等变网络回归形状和姿态，所有模块均端到端训练。我们的新型等变姿态回归模块利用自注意力层的置换等变特性保持旋转等变性。实验结果表明，ArtEq可泛化至训练中未见的姿态，在无需优化精化步骤的情况下，比现有最优方法性能提升74.5%。此外，与竞争方法相比，本方法推理速度快三个数量级以上，参数量减少97.3%。代码与模型将在https://arteq.is.tue.mpg.de 公开供研究使用。