Weakly Supervised Semantic Segmentation Based on Co-segmentation

Tong Shen, Guosheng Lin, Lingqiao Liu, Chunhua Shen and Ian Reid

Abstract

Training a Fully Convolutional Network (FCN) for semantic segmentation requires a large number of masks with pixel level labelling, which involves a large amount of human labour and time for annotation. In contrast, web images and their image-level labels are much easier and cheaper to obtain. In this work, we propose a novel method for weakly supervised semantic segmentation with only image-level labels. The method utilizes the internet to retrieve a large number of images and uses a large scale co-segmentation framework to generate masks for the retrieved images. We first retrieve images from search engines, e.g. Flickr and Google, using semantic class names as queries, e.g. class names in the dataset PASCAL VOC 2012. We then use high quality masks produced by co-segmentation on the retrieved images as well as the target dataset images with image level labels to train segmentation networks. We obtain an IoU score of 56.

Session

Posters

Files

Paper (PDF)

DOI

10.5244/C.31.17
https://dx.doi.org/10.5244/C.31.17

Citation

Bibtex

            @inproceedings{BMVC2017_17,
                title={Weakly Supervised Semantic Segmentation Based on Co-segmentation},
                author={Tong Shen, Guosheng Lin, Lingqiao Liu, Chunhua Shen and Ian Reid},
                year={2017},
                month={September},
                pages={17.1-17.12},
                articleno={17},
                numpages={12},
                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.31.17},
                isbn={1-901725-60-X},
                url={https://dx.doi.org/10.5244/C.31.17}
            }