Saliency detection by conditional generative adversarial network

Xiaoxu Cai, Hui Yu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

433 Downloads (Pure)


Detecting salient objects in images has been a fundamental problem in computer vision. In recent years, deep learning has shown its impressive performance in dealing with many kinds of vision tasks. In this paper, we propose a new method to detect salient objects by using Conditional Generative Adversarial Network (GAN). This type of network not only learns the mapping from RGB images to salient regions, but also learns a loss function for training the mapping. To the best of our knowledge, this is the first time that Conditional GAN has been used in salient object detection. We evaluate our saliency detection method on 2 large publicly available datasets with pixel accurate annotations. The experimental results have shown the significant and consistent improvements over the state-of-the-art methods on a challenging dataset, and the testing speed is much faster.
Original languageEnglish
Title of host publicationProceedings Volume 10615, Ninth International Conference on Graphic and Image Processing (ICGIP 2017)
PublisherSPIE Press
Publication statusPublished - 10 Apr 2018
Event9th International Conference on Graphic and Image Processing: ICGIP 2017 - Qingdao, China
Duration: 14 Oct 201716 Oct 2017


Conference9th International Conference on Graphic and Image Processing
Internet address


  • saliency detection
  • deep learning
  • Generative Adversarial Network
  • CGAN


Dive into the research topics of 'Saliency detection by conditional generative adversarial network'. Together they form a unique fingerprint.

Cite this