3D human pose estimation from 2D detections is an ill-posed problem because multiple solutions may exist due to the inherent ambiguity and occlusion. In this paper, we propose a novel graph-based mixture density network (GMDN) to tackle the 2D-to-3D human pose estimation problem. We formulate the 2D joint locations of the human body as a graph, and thus the pose estimation task can be redefined as a graph regression problem. Additionally, we present a novel graph convolutional operation with the incorporation of structural knowledge about human body configurations to assist with reasoning of the structural relations implied in the human bodies. Furthermore, we employ mixture density networks to formulate the 3D human poses as a multimodal distribution. The presented GMDN is lightweight with only 0.30M parameters, and the experimental results demonstrate that it achieves state-of-the-art performance.

通过2D检测进行3D人体姿势估计是一个不适的问题，因为由于固有的歧义和遮挡，可能存在多个解决方案。在本文中，我们提出了一种新颖的基于图形的混合密度网络（GMDN），以解决2D到3D人体姿势估计问题。我们将人体的二维关节位置公式化为图形，因此姿势估计任务可以重新定义为图形回归问题。此外，我们提出了一种新颖的图卷积运算，其中包含有关人体构造的结构知识，以协助推理隐含在人体中的结构关系。此外，我们采用混合密度网络将3D人体姿势公式化为多峰分布。所展示的GMDN是轻量级的，只有0.30M参数，