Model construction method, motion capture mark generation method, system, equipment and medium

The invention provides a model construction method, a motion capture mark generation method, a system, equipment and a medium, and the model construction method comprises the steps: obtaining a first spatial deformation model which is used for carrying out the affine transformation of an image; obtaining a second spatial deformation model, wherein the second spatial deformation model is used for performing pixel-level change on the image; the first space deformation model and the second space deformation model are integrated to obtain an integrated space deformation model, and the integrated space deformation model can perform affine transformation and pixel-level nonlinear transformation on an image in sequence to obtain a Congealing image. According to the method, the integrated space deformation model is constructed, so that the integrated space deformation model can perform affine transformation and pixel-level nonlinear transformation on the image in sequence to obtain the Congealing image, thereby avoid