Medical-Human-Pose dataset

The Medical-Human-Pose dataset contains 3752 depth images of people standing, sitting, lying, and falling in medical scenarios (laboratory simulations). For each image, the location of 15 human body parts are labeled with 3-dimensional (x,y,z) coordinates relative to the sensor position, the 4 vertex positions of the human target surround box are labeled as two-dimensional (x,y) coordinates.Medical-Human-Pose 数据集（Medical-Human-Pose Dataset）包含3752张医疗场景（实验室模拟环境）下的深度图像，图像涵盖人体站立、坐姿、躺卧及跌倒四种姿态。针对每张图像，15个人体部位的位置均以相对于传感器位置的三维（x,y,z）坐标标注；人体目标包围框的四个顶点位置则以二维（x,y）坐标完成标注。