Point attention network for semantic segmentation of 3D point clouds

作者：

Highlights：

• A point attention network that learns rich local shape features and their contextual correlations for 3D point cloud semantic segmentation.

• A Local Attention-Edge Convolution which constructs a local graph based on the neighborhood points searched in multi-directions.

• A point-wise spatial attention module which captures long-range spatial contextual features contributing to more precise semantic information.

• Extending the U-shaped network to incorporate the proposed Local Attention-Edge Convolution layer and point-wise spatial attention module.

摘要

•A point attention network that learns rich local shape features and their contextual correlations for 3D point cloud semantic segmentation.•A Local Attention-Edge Convolution which constructs a local graph based on the neighborhood points searched in multi-directions.•A point-wise spatial attention module which captures long-range spatial contextual features contributing to more precise semantic information.•Extending the U-shaped network to incorporate the proposed Local Attention-Edge Convolution layer and point-wise spatial attention module.

论文关键词：Semantic segmentation,3D point cloud,Point attention network,Deep learning

论文评审过程：Received 27 September 2019, Revised 23 April 2020, Accepted 13 May 2020, Available online 17 May 2020, Version of Record 7 June 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107446