Boundarymix: Generating pseudo-training images for improving segmentation with scribble annotations

作者：

Highlights：

• BoundaryMix is proposed for scribble-supervised semantic segmentation.

• BoundaryMix supplements the missing boundary information of scribble annotations by generating pseudo training images and annotations.

• Scribble are used to annotate remote sensing images and show that scribble annotation is also suitable for different scenarios.

• Experiments on PASCAL VOC and POTSDAM datasets show that BoundaryMix almost closes the gap between weakly-supervised and fully-supervised semantic segmentation.

摘要

•BoundaryMix is proposed for scribble-supervised semantic segmentation.•BoundaryMix supplements the missing boundary information of scribble annotations by generating pseudo training images and annotations.•Scribble are used to annotate remote sensing images and show that scribble annotation is also suitable for different scenarios.•Experiments on PASCAL VOC and POTSDAM datasets show that BoundaryMix almost closes the gap between weakly-supervised and fully-supervised semantic segmentation.

论文关键词：Weakly-supervised segmentation,Scribble,Boundary mix

论文评审过程：Received 3 July 2020, Revised 26 January 2021, Accepted 28 February 2021, Available online 26 March 2021, Version of Record 16 April 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.107924