Unsupervised CNN-Based Co-saliency Detection with Graphical Optimization

Kuang Jui Hsu, Chung Chi Tsai, Yen Yu Lin, Xiaoning Qian, Yung Yu Chuang

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

8 Scopus citations


In this paper, we address co-saliency detection in a set of images jointly covering objects of a specific class by an unsupervised convolutional neural network (CNN). Our method does not require any additional training data in the form of object masks. We decompose co-saliency detection into two sub-tasks, single-image saliency detection and cross-image co-occurrence region discovery corresponding to two novel unsupervised losses, the single-image saliency (SIS) loss and the co-occurrence (COOC) loss. The two losses are modeled on a graphical model where the former and the latter act as the unary and pairwise terms, respectively. These two tasks can be jointly optimized for generating co-saliency maps of high quality. Furthermore, the quality of the generated co-saliency maps can be enhanced via two extensions: map sharpening by self-paced learning and boundary preserving by fully connected conditional random fields. Experiments show that our method achieves superior results, even outperforming many supervised methods.
Original languageAmerican English
Title of host publicationLecture Notes in Computer Science
PublisherSpringer Verlag
Number of pages17
ISBN (Print)9783030012274
StatePublished - 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11209 LNCS


  • Co-saliency detection
  • Convolutional neural networks
  • Deep learning
  • Graphical model
  • Unsupervised learning


Dive into the research topics of 'Unsupervised CNN-Based Co-saliency Detection with Graphical Optimization'. Together they form a unique fingerprint.

Cite this