DeepSaliency: Multi-Task Deep Neural Network Model for Salient Object Detection

A key problem in salient object detection is how to effectively model the semantic properties of salient objects in a data-driven manner. In this paper, we propose a multi-task deep saliency model based on a fully convolutional neural network with global input (whole raw images) and global output (whole saliency maps). In principle, the proposed saliency model takes a data-driven strategy for encoding the underlying saliency prior information, and then sets up a multi-task learning scheme for exploring the intrinsic correlations between saliency detection and semantic image segmentation. Through collaborative feature learning from such two correlated tasks, the shared fully convolutional layers produce…

• NN
  National Natural Science Foundation of China

  Awards: 61528204, 61472353, U1509206
• NK
  National Key Research and Development Program of China

  Awards: 2015CB352302, 2012CB316400
• FR
  Fundamental Research Funds for the Central Universities
• DO
  Division of Information and Intelligent Systems

  Awards: 1152576, 1350521, 1218156, 1149783