Adversarial network integrating dual attention and sparse representation for semi-supervised semantic segmentation
作者：
Highlights：
• A semi-supervised GAN framework for semantic segmentation is proposed.
• The dual attention is proposed to model the global and local semantic dependencies.
• The sparse representation module is adopted to further improve the performance.
• The focal attention map is proposed to enhance the robustness of the model.
• Extensive experiments are conducted to verify the effectiveness of our method.
摘要
•A semi-supervised GAN framework for semantic segmentation is proposed.•The dual attention is proposed to model the global and local semantic dependencies.•The sparse representation module is adopted to further improve the performance.•The focal attention map is proposed to enhance the robustness of the model.•Extensive experiments are conducted to verify the effectiveness of our method.
论文关键词：Semi-supervised learning,Focal attention map,Dual attention,Sparse representation,Adversarial network
论文评审过程：Received 4 March 2021, Revised 28 June 2021, Accepted 29 June 2021, Available online 12 July 2021, Version of Record 12 July 2021.
论文官网地址：https://doi.org/10.1016/j.ipm.2021.102680
