Abstract
In this paper a new Sparse Tensor Auto-Encoder (STAE) model is proposed to learn a latent and discriminative feature representations for saliency detection. By formulating the background patches as holistic high-dimensional tensors and learning multi-dimensional dictionary to code image patches, the coding error can precisely reveal the difference between salient object and background. Then a saliency map can be derived by a subsequent refinement of the representation errors of patches via image segmentation. Several benchmark datasets are used to verify the effectiveness of the proposed method. The results show that the proposed STAE can accurately locate saliency region and outperform its counterparts.
Funding Information
  • National Natural Science Foundation of China (61771376, 91438103, 91438201, 61906145, 61771380, 61703328, U1730109, U1701267)
  • Equipment pre-research project of the 13th Five-Years Plan (6140312010103, 6141A020223, 6140137050206, 6141B06160301, 6141B07090102, 414120101026)
  • Major Research Plan in Shaanxi Province of China (017ZDCXL-GY-03-02, 2017ZDXM-GY-103)
  • Foundation of the State Key Laboratory of Complex Electromagnetic Environment Effects (2017K0202B, 2019Z0101B, 2019K0203B, 2018K0101B)
  • Science Basis Research Program in Shaanxi Province of China (2017JM6086, 2019JQ-663, 16JK1823)
  • Key Scientific and Technological Innovation Team of Shaanxi Province (S2020-ZC-TD-0084)

This publication has 25 references indexed in Scilit: