Semantic Segmentation with Reverse Attention

Qin Huang, Chihao Wu, Chunyang Xia, Ye Wang and C.-C. Jay Kuo

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 mean IoU score (48.1%) for the challenging PASCAL-Context dataset.

Session

Posters

Files

Paper (PDF)
Supplementary (PDF)

DOI

10.5244/C.31.18
https://dx.doi.org/10.5244/C.31.18

Citation

Bibtex

            @inproceedings{BMVC2017_18,
                title={Semantic Segmentation with Reverse Attention},
                author={Qin Huang, Chihao Wu, Chunyang Xia, Ye Wang and C.-C. Jay Kuo},
                year={2017},
                month={September},
                pages={18.1-18.13},
                articleno={18},
                numpages={13},
                booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
                publisher={BMVA Press},
                editor={Tae-Kyun Kim, Stefanos Zafeiriou, Gabriel Brostow and Krystian Mikolajczyk},
                doi={10.5244/C.31.18},
                isbn={1-901725-60-X},
                url={https://dx.doi.org/10.5244/C.31.18}
            }