Abstract: To address the issue that it is difficult to effectively maintain the detail information of original shapes for traditional point cloud restoration methods, a two-branch point cloud completion network is proposed which combines shape structure recovery branch and local detail compensation branch. For the shape structure recovery branch, its encoder performs feature transformation on the missing point cloud data to overcome the rotational invariance of the 3d shapes, and solves the disorder problem of the point cloud by using the maximum pooling operation, which can generate the feature codeword of the input shape by adopting the multi-layer perceptron. In order to compensate shape details of the coarse completion results, the local detail compensation branch can learn the geometric features of the underlying shapes through hierarchical feature learning and multi-level feature fusion for different dimensional features extracted from the encoder, so as to effectively recover the missing point cloud data and retain the original shape details. Finally, the point cloud data obtained by these two branches will be stitched and fused, and then the farthest point resampling is iteratively performed to obtain the final point cloud completion results. Compared with the completion results on ShapeNet dataset, the average CD error and EMD error of our proposed network are 16%~29% and 19%~65% lower than that of the existing networks respectively, whilst the average CD error and EMD error are 6%~41% and 31%~59% lower than that of the existing network respectively if testing on the ModelNet dataset. The proposed two-branch completion network can repair the overall structure of the underlying shape and effectively recover its shape details to generate a complete point cloud model with uniform distribution of sampling points. The network is also robust to model noise and different degrees of missing data.