Abstract: This presentation proposes a method, based on data fusion, to obtain 3D features of polyhedra using co-registered range and intensity images. First, feature points and edges of the candidate planes of the objects in the intensity and range images are acquired by analysing the intensity and range variation. Then, all the 3D candidate venices, edges and planes in the range image can be obtained using the correspondence of the two co-registered images. Next, the candidate planes are verified by computing and analysing the curvatures and normals at some feature points and edges on the candidate planes in the range image. Finally, the verified candidates are regarded as actual planes of the sensed object, and are used to construct a hierarchical representation of the object. Experiment results are given to show the feasibility of the proposed approach.