There has been a recent surge of interest in introducing transformers to 3D human pose estimation (HPE) due to their powerful capabilities in modeling long-term dependencies. However, existing transformer-based methods treat body joints as equally important inputs and ignore the prior knowledge of human skeleton topology in the self-attention mechanism. To tackle this issue, in this paper, we propose a Pose-Oriented Transformer (POT) with uncertainty guided refinement for 3D HPE. Specifically, we first develop novel pose-oriented self-attention mechanism and distance-related position embedding for POT to explicitly exploit the human skeleton topology. The pose-oriented self-attention mechanism explicitly models the topological interactions between body joints, whereas the distance-related position embedding encodes the distance of joints to the root joint to distinguish groups of joints with different difficulties in regression. Furthermore, we present an Uncertainty-Guided Refinement Network (UGRN) to refine pose predictions from POT, especially for the difficult joints, by considering the estimated uncertainty of each joint with uncertainty-guided sampling strategy and self-attention mechanism. Extensive experiments demonstrate that our method significantly outperforms the state-of-the-art methods with reduced model parameters on 3D HPE benchmarks such as Human3.6M and MPI-INF-3DHP

翻译：近年来，由于Transformer在建模长程依赖关系方面的强大能力，将其引入三维人体姿态估计（HPE）的研究兴趣激增。然而，现有的基于Transformer的方法将身体关节视为同等重要的输入，并在自注意力机制中忽略了人体骨骼拓扑结构的先验知识。为解决这一问题，本文提出了一种面向姿态的Transformer（POT），并结合不确定性引导精化用于三维HPE。具体而言，我们首先为POT开发了新颖的面向姿态的自注意力机制和距离相关位置嵌入，以显式利用人体骨骼拓扑结构。面向姿态的自注意力机制显式建模了身体关节之间的拓扑交互，而距离相关位置编码则对关节到根关节的距离进行编码，以区分回归难度不同的关节组。此外，我们提出了一种不确定性引导精化网络（UGRN），通过考虑每个关节的估计不确定性，采用不确定性引导的采样策略和自注意力机制，对POT的姿态预测进行精化，特别是针对困难关节。大量实验表明，我们的方法在三维HPE基准（如Human3.6M和MPI-INF-3DHP）上，以更少的模型参数显著优于当前最先进的方法。