Point cloud classification with deep normalized Reeb graph convolution

作者：

Highlights：

• Deep graph convolutions on structured high-level Reeb graph on point-clouds.

• Graph Normalization and Normalized Similarity Distance to identify graph structure.

• Information fusion from both local and global contexts on point-clouds.

• Trained solely on canonical poses but generalizes to arbitrarily rotated objects.

摘要

•Deep graph convolutions on structured high-level Reeb graph on point-clouds.•Graph Normalization and Normalized Similarity Distance to identify graph structure.•Information fusion from both local and global contexts on point-clouds.•Trained solely on canonical poses but generalizes to arbitrarily rotated objects.

论文关键词：Reeb graph,Point cloud,Graph normalization

论文评审过程：Received 14 July 2020, Revised 2 November 2020, Accepted 8 December 2020, Available online 13 December 2020, Version of Record 2 January 2021.

论文官网地址：https://doi.org/10.1016/j.imavis.2020.104092