Learning graph affinities for spectral graph-based salient object detection

详细信息    查看全文

• 作者：Ç ; ağlar Aytekin^a ; ^{caglar.aytekin@tut.fi}Author Vitae ; Alexandros Iosifidis^a ; ^{alexandros.iosifidis@tut.fi}Author Vitae ; Serkan Kiranyaz^b ; ^{mkiranyaz@qu.edu.qa}Author Vitae ; Moncef Gabbouj^a ; ^{moncef.gabbouj@tut.fi}Author Vitae
• 关键词：Salient object detection ; Graph affinities ; Spectral graph theory
• 刊名：Pattern Recognition
• 出版年：2017
• 出版时间：April 2017
• 年：2017
• 卷：64
• 期：Complete
• 页码：159-167
• 全文大小：543 K
• 卷排序：64

文摘

We propose a method for learning graph affinities for salient object detection. We employ CKNs with a global inference layer EQCUT for salient object detection. We provide backpropagation rules of both CKN and EQCUT for parameter learning. The proposed system is trained end-to-end for performance enhancement.