Improving deep learning on point cloud by maximizing mutual information across layers

作者：

Highlights：

• A novel MMI-based method is proposed for improving learning the point cloud representation without additional data augmentation.

• The proposed maximizing mutual information (MMI) module can be flexibly applied in various point cloud networks.

• A novel MI-based loss function considers both local and global semantic information of hierarchical point cloud features.

• Experimental results demonstrate the efficacy and board applicability of our method.

摘要

•A novel MMI-based method is proposed for improving learning the point cloud representation without additional data augmentation.•The proposed maximizing mutual information (MMI) module can be flexibly applied in various point cloud networks.•A novel MI-based loss function considers both local and global semantic information of hierarchical point cloud features.•Experimental results demonstrate the efficacy and board applicability of our method.

论文关键词：Deep learning,3D vision,Point clouds,Mutual information

论文评审过程：Received 2 March 2022, Revised 25 June 2022, Accepted 6 July 2022, Available online 8 July 2022, Version of Record 15 July 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.108892