U2-Net: Going deeper with nested U-structure for salient object detection

作者：

Highlights：

• A novel ReSidual U-block (RSU) is designed to capture multi-scale deep features.

• A nested U-structure, called U2-Net, that uses RSU is developed for salient object detection.

• Both large (176.3 MB) and small (4.7 MB) instances of U2-Net get competitive results.

摘要

•A novel ReSidual U-block (RSU) is designed to capture multi-scale deep features.•A nested U-structure, called U2-Net, that uses RSU is developed for salient object detection.•Both large (176.3 MB) and small (4.7 MB) instances of U2-Net get competitive results.

论文关键词：Salient object detection,Convolutional neural network,Network architecture design,Nested U-structure,Multi-scale feature extraction

论文评审过程：Received 27 November 2019, Revised 6 April 2020, Accepted 28 April 2020, Available online 16 May 2020, Version of Record 16 May 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107404