At present, clothing structure design technology is experiencing a transformation from two-dimensional to three-dimensional, among which, three-dimensional clothing modeling technology is an important part of three-dimensional clothing structure design research. In this paper, on the basis of the SCAPE method, the global mapping method is adopted to obtain the relationship between human body size parameters and PCA weights, and the deformation law of the human body network is deduced by learning the deformation matrix to realize the establishment of personalized 3D human body model based on personalization. After processing the human polygonal model, the NURBS surface is used to create a 3D human body model that indicates smoothness and high realism. Using the Kinect 3D body scanner, model data is collected to construct a 3D anthropometric database that provides data support for subsequent clothing design. Define the simulation state of fabric organization in the computer, construct the geometric structure model of fabric, and model the clothing fabric. Several mainstream algorithms are selected and compared with the algorithm proposed in this paper. The running time of the algorithm proposed in this paper is reduced by 0.3596s and 0.3022s compared to the geometric model method and the constraint model method, which reflects the superiority of the algorithm. The fabric thickness generated by this paper’s algorithm ranges from 1.94-1.99, and the fabric recovery rate in three minutes is more than 98%, the fabric performance is better, and the three-dimensional garment structure design proposed in this paper is effective.