This paper presents a new approach for the classification and retrieval of three-dimensional images and models from databases. A set of retrieval algorithms is introduced. These algorithms are content-based, meaning that the input is not made out of keywords but of three-dimensional models. Tensors of inertia, distribution of normal vectors, distribution of cords and multiresolution analysis are used to describe each model. Scale, shape or color or any combination of these parameters can search the database. A user friendly interface makes the retrieval operation simple and intuitive and make it possible to edit reference models according to the specifications of the user. Experimental results using a database of more than 1000 VRML models are presented.