3D姿态估计的评价指标MPJPE及其变种

MPJPE是“Mean Per Joint Position Error”，即“平均（每）关节位置误差”。

MPJPE常常用于3D Human Pose Estimation算法的评价指标，这个指标越小则可认为这个3D人体姿态估计算法越好。

看一下公式就能一目了然：

截图来自论文：http://vision.imar.ro/human3.6m/pami-h36m.pdf

用白话描述就是：计算预测关节点与对应GT关节点的L2距离的平均值。

MPJPE通常被称为Protocol #1，常用的还有Protocol #2。此外还有不太常用的Protocol #3。

Protocol #2是P-MPJPE，先经过旋转、对齐等变换再进行MPJPE；而Protocol #3是N-MPJPE，先进行规模对齐再进行MPJPE。

咱们来看一段描述：

Protocol 1 is the mean per-joint position error (MPJPE) in millimeters which is the mean Euclidean distance between predicted joint positions and ground-truth joint positions and follows [29, 53, 59, 34, 41]. Protocol 2 reports the error after alignment with the ground truth in translation, rotation, and scale (P-MPJPE) [34, 50, 10, 40, 56, 16]. Protocol 3 aligns predicted poses with the ground-truth only in scale (N-MPJPE) following [45] for semi-supervised experiments.

描述来自论文：https://arxiv.org/pdf/1811.11742.pdf