Multi-layer graph constraints for interactive image segmentation via game theory
详细信息 查看全文
- 作者：Tao Wang wangtaoatnjust@163.com" class="auth_mail" title="E-mail the corresponding author 670101389@qq.com" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Quansen Sun ; sunquansen@njust.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Zexuan Ji ; jizexuan@njust.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Qiang ChenAuthor Vitae ; Peng FuAuthor Vitae
- 关键词：Image segmentation ; Superpixel ; Multi-layer graph ; Nonparametric learning ; Game theory
- 刊名：Pattern Recognition
- 出版年：2016
- 出版时间：July 2016
- 年：2016
- 卷：55
- 期：Complete
- 页码：28-44
- 全文大小：13738 K
文摘
- •
Multi-layer graph constraints are utilized for the interactive image segmentation.
- •
Labeling information is introduced into the conventional pixel–superpixel combinational model.
- •
The optimization based on game theory is proposed for the combinational energy functions.
- •
The proposed method can obtain better performance than the state-of-the-art methods.