Parametric Curves And Surfaces In The Garment Cad Modeling Studies

Garment CAD (GCAD) is the information technology which is widely used in the garment industry. It has the great practical significance for improving the production efficiency and the product quality, more effectively saving the raw material resource, shortening the garment production and the sale cycle, reducing the production cost, etc. In GCAD, we normally use curves or surfaces for 2D structure design or 3D effect design. Different curves or surfaces we choose determine the style of the GCAD software. Whether the curves or surfaces are proper or not has the direct relationship with the quality of the garment design, so the curve and surface modeling study on GCAD has an important research and application value.This paper does the research on common parametric curve and surface modeling in Computer Graphics and applies these methods for garment CAD. The main content is as follows:Section 1: It analyzes the current research state and development trend in the garment CAD system.Section 2: It describes common parametric curve and surface modeling and analyzes their advantages and disadvantages in garment CAD.Section 3: It does the research on cubic Bézier spline curve (surface) construction and its application in GCAD. The computation and analysis shows that cubic Bézier spline curve obtains good connecting results for solving the problem that the curve is tangent to the control polygon in the garment pattern design process. Furthermore, the curve's shape can be modified by the adjusting parameters. Then, bi-cubic Bézier spline surface is constructed by extending the cubic Bézier spline curve method to 3D space. The spline surface is smooth and has a good overall effect for the garment modeling. Moreover, it can modify the surface to create the folding effect for the 3D garment modeling, which is very useful in the 3D garment design. Section 4: It gives the non-photorealistic rendering research on 3D garment mesh models. We firstly propose a new estimation method on discrete differential geometry properties of triangular mesh models. This method constructs the local patch for fitting each vertex and its neighboring vertices in a mesh model by the weighted bi-cubic Bézier surface. Then, it estimates normal vector and curvature by this local fitting patch. It also defines the principal geodesic torsion on each vertex and estimates it by above fitting patch. Finally, it creates the gray diagram of the mesh model by the linear combination of the estimated curvature and torsion, which is applied for the non-photorealistic rendering of 3D garment mesh models.Section 5: It has the research on the matching between 3D body model and garment model. Firstly, we fit each vertex and its local neighborhood of the clothing and human models by constructing the weighted bi-cubic Bézier surface, and then estimate the discrete curvature and torsion on each vertex. All of the feature points can be extracted by the feature extraction method. The matching feature points of two mesh models are extracted by constructing new matching function according to curvatures and torsions. The coordinate transformation matrix is calculated by minimizing the distance function. After the rough match result is obtained, the clothing model is scaled and matched again to achieve the accurate matching and the 3D clothing fitting effect.In the end of this paper, it summarizes the current work and gives the development prospects of this research.