Deep-Dense Conditional Random Fields for Object Co-Segmentation
Abstract
We address the problem of object co-segmentation in images. Object co-segmentation aims to segment common objects in images and has promising applications in AI agents. We solve it by proposing a co-occurrence map, which measures how likely an image region belongs to an object and also appears in other images. The co-occurrence map of an image is calculated by combining two parts: objectness scores of image regions and similarity evidences from object proposals across images. We introduce a deep-dense conditional random field framework to infer co-occurrence maps. Both similarity metric and objectness measure are learned end-to-end in a single deep network. We evaluate our method on two benchmarks and achieve competitive performance.
Cite
Text
Yuan et al. "Deep-Dense Conditional Random Fields for Object Co-Segmentation." International Joint Conference on Artificial Intelligence, 2017. doi:10.24963/IJCAI.2017/471Markdown
[Yuan et al. "Deep-Dense Conditional Random Fields for Object Co-Segmentation." International Joint Conference on Artificial Intelligence, 2017.](https://mlanthology.org/ijcai/2017/yuan2017ijcai-deep/) doi:10.24963/IJCAI.2017/471BibTeX
@inproceedings{yuan2017ijcai-deep,
title = {{Deep-Dense Conditional Random Fields for Object Co-Segmentation}},
author = {Yuan, Ze-Huan and Lu, Tong and Wu, Yirui},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2017},
pages = {3371-3377},
doi = {10.24963/IJCAI.2017/471},
url = {https://mlanthology.org/ijcai/2017/yuan2017ijcai-deep/}
}