DSANet: Dilated spatial attention for real-time semantic segmentation in urban street scenes

作者：

Highlights：

• Real time semantic segmentation for road scene understanding.

• Follow two branch network strategy.

• Utilize dilated spatial attention and channel attention.

• Use supervision to ease the loss propagation.

摘要

•Real time semantic segmentation for road scene understanding.•Follow two branch network strategy.•Utilize dilated spatial attention and channel attention.•Use supervision to ease the loss propagation.

论文关键词：Street scene understanding,Deep learning,Lightweight convolutional,Neural network,Real-time semantic segmentation,Spatial attention

论文评审过程：Received 7 December 2020, Revised 26 March 2021, Accepted 26 March 2021, Available online 29 April 2021, Version of Record 14 June 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115090