Distinguishing foreground and background alignment for unsupervised domain adaptative semantic segmentation
作者：
Highlights：
• We use self-supervised learning to generate pseudo-labels for the target domain.
• We distinguish and align the foreground and background classes.
• We use parallel attention module to capture the space and channel information.
• We add focal loss to the overall loss to reduce the impact of class imbalance.
摘要
•We use self-supervised learning to generate pseudo-labels for the target domain.•We distinguish and align the foreground and background classes.•We use parallel attention module to capture the space and channel information.•We add focal loss to the overall loss to reduce the impact of class imbalance.
论文关键词：Semantic segmentation,Self-supervised learning,pseudo labels,Attention mechanism,Focal loss
论文评审过程：Received 18 January 2022, Revised 16 May 2022, Accepted 20 June 2022, Available online 25 June 2022, Version of Record 1 July 2022.
论文官网地址：https://doi.org/10.1016/j.imavis.2022.104513
