The current object segmentation algorithms don′t work well in cluttered point c ouds.To solve the problem of incomplete or over segmentation of clutered point c loud in the research of indoor 3D object recognition,we introduce a novel approa ch for segmenting the cluttered point cloud on the desktop by using bilateral sy mmetry constraints.By extracting and matching surface normal boundary curves in the point cloud to be segmented,detecting 3D bilateral symmetryin scene and crea ting a set of symmetric hypotheses,the segmentation of cluttered point cloud is finished.We evaluate our approach on a dataset of 3D pintcloud of Maryland.And c omp ared with regional growth algorithm and LCCP algorithm,this algorithm works bett er in average scoreevaluation parameters.The experimental results show that the proposed algorithm has the best segmentation effect,complete segmentation under vision and the best evaluation index.It′s superior to the current manstream seg mentation algorithms.