This paper presents a consumer-oriented design approach to determining the optimal form design of character toys that optimal matches a given set of product images perceived by consumers. 179 representative character toys and seven design form elements of character toys are identified as samples in an experimental study to illustrate how the consumer-oriented design approach works. The consumer-oriented design approach is based on the process of Kansei Engineering using neural networks (NNs). Nine NN models are built with different momentum, learning rate, and hidden neurons in order to examine how a particular combination of form elements matches the desirable product images. The NN models can be used to construct a form design database for supporting form design decisions in a new character toy design. The result provides useful insights that help product designers best meet consumers' specific feelings and expectations.