blender_lifting is a Blender addon that implements the functionality from Denis Tome’ research paper Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image, which uses machine learning to estimate a 3D pose from a 2D image. More information can be found about this can be found in its GitHub repository and its associated research paper.

This addon will take an image of a person, estimate its pose and create a rigged skeleton in Blender in that pose, as illustrated in the example below.

Source code and addon available on GitHub