Digital cameras are often used as colorimeters where the task is to predict as accurate colorimetric values of an object as possible from camera responses of the object. It has been widely recognized that one of the factors that makes color prediction less straightforward is metamerism.This paper addresses a novel color prediction method that utilizes a population of spectral reflectance samples of real objects based on which the effect of metamerism is taken into consideration. The results of computer simulations show that the proposed method is able to achieve higher performance than the conventional matrix-based color prediction method.