PCANet: Pyramid convolutional attention network for semantic segmentation

作者：

Highlights：

• A novel module named PAA to learn long-term dependency and embed location information into features.

• Firstly fuses both channel and spatial attention mechanism and applies this mechanism in features fusion.

• Combine both of PAA module and CAR module to construct our PCANet for semantic segmentation.

摘要

•A novel module named PAA to learn long-term dependency and embed location information into features.•Firstly fuses both channel and spatial attention mechanism and applies this mechanism in features fusion.•Combine both of PAA module and CAR module to construct our PCANet for semantic segmentation.

论文关键词：Non-local module,Atrous convolution,Attention mechanism,Semantic segmentation

论文评审过程：Received 24 June 2020, Revised 28 July 2020, Accepted 31 July 2020, Available online 7 August 2020, Version of Record 26 August 2020.

论文官网地址：https://doi.org/10.1016/j.imavis.2020.103997