With diversification of consumer taste, appearance shape together with functionality contributes to the appeal of a product vastly. Concept design and industrial design therefore serve as an important process in product development. These designs are difficult to perform based on theoretical backing, since appearance shape design is a creative activity which depends on a designer’s aesthetic sense strongly. When embodying a product shape, naturally design is determined not only by a designer’s sensitivity but by use and function of a product as well. It is also important to investigate designs desired by consumers, and reflect all of this in the product design. The ability to predict consumer taste trends therefore greatly aids product design. In this research, the prototype models of a product in trend every year were made by multiplying weights according to the number of a product sold in the past to calculate that the rate of exaggeration of prototype models of each year to all whole prototype models. The straight extrapolation of the Spline method was applied to the exaggeration vector, and the technique of predicting shapes preferred by consumers in the near future using that method was proposed. Moreover the eigenspace method was applied to similar product shapes to propose the technique of grasping the features of shape for every year by computing the eigenvalue and eigenvector of the coordinates of the points of the shapes as well as the technique of predicting shapes which consumers will prefer in the near future by using the Linear function of Moving Least Square method.
Superresolution of 3-D computational integral imaging based on moving least square method
