CODH++: Macro-semantic differences oriented instance segmentation network

作者：

Highlights：

• We exposed the flaws of the two-stage instance segmentation method.

• A novel two stage instance segmentation paradigm is presented.

• An ablation study illustrates the effect of different elements of the framework.

• State-of-the-art methods are outperformed in both parameters and FLOPs.

摘要

•We exposed the flaws of the two-stage instance segmentation method.•A novel two stage instance segmentation paradigm is presented.•An ablation study illustrates the effect of different elements of the framework.•State-of-the-art methods are outperformed in both parameters and FLOPs.

论文关键词：Instance segmentation,Macro-semantic morphological,Mask head,Long-range dependencies

论文评审过程：Received 7 August 2021, Revised 5 April 2022, Accepted 6 April 2022, Available online 12 April 2022, Version of Record 28 April 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117198