Semantic Segmentation with Reverse Attention
Abstract
Recent development in fully convolutional neural network enables efficient end-to-end learning of semantic segmentation. Traditionally, the convolutional classifiers are taught to learn the representative semantic features of labeled semantic objects. In this work, we propose a reverse attention network (RAN) architecture that trains the network to capture the opposite concept (i.e., what are not associated with a target class) as well. The RAN is a three-branch network that performs the direct, reverse and reverse-attention learning processes simultaneously. Extensive experiments are conducted to show the effectiveness of the RAN in semantic segmentation. Being built upon the DeepLabv2-LargeFOV, the RAN achieves the state-of-the-art mIoU score (48.1%) for the challenging PASCAL-Context dataset. Significant performance improvements are also observed for the PASCAL-VOC, Person-Part, NYUDv2 and ADE20K datasets.
Cite
@article{arxiv.1707.06426,
title = {Semantic Segmentation with Reverse Attention},
author = {Qin Huang and Chunyang Xia and Chihao Wu and Siyang Li and Ye Wang and Yuhang Song and C. -C. Jay Kuo},
journal= {arXiv preprint arXiv:1707.06426},
year = {2017}
}
Comments
accepted for oral presentation in BMVC 2017