Upsampling Autoencoder for Self-Supervised Point Cloud Learning

While a large number of supervised learning methods have been proposed to handle the unordered point clouds and achieved remarkable success, their performance is limited to the costly data annotation. In this work, we propose a novel self-supervised pre-training model for point cloud learning without human annotations, which relies solely on upsampling operation to perform feature learning of point cloud in an effective manner. The key observation of our approach is that upsampling operation encourages the network to capture both high-level semantic information and low-level geometric information of the point cloud. As a result, the downstream tasks such as classification and segmentation will benefit from the pre-trained model. And our method outperforms previous methods in shape classification, part segmentation.