Abstract: A new method of modeling plants with large leaves from 3D point clouds is presented.The algorithm consists of two parts:points clustering and leaf reconstruction.Firstly,the points are clustered by computing distances between them.Secondly,based on the characteristics of leaves,whose point clouds are approximately on a plane,the points are clustered.And last a leaf is modeled by fitting the clustered point clouds to a leaf model.The experiment results show that the algorithm is stable,and the proposed method can be used to reconstruct the plants with large leaves.