Structural Saliency: The Detection of Globally Salient Structures Using a Locally Connected Network

Abstract

Certain salient structures in images attract our immediate attention without requiring a systematic scan. We present a method for computing saliency by a simple iterative scheme, using a uniform network of locally connected processing elements. The network uses an optimization approach to produce a "saliency map," a representation of the image emphasizing salient locations. The main properties of the network are: (i) the computations are simple and local, (ii) globally salient structures emerge with a small number of iterations, and (iii) as a by-product of the computations, contours are smoothed and gaps are filled in.

Cite

Text

Shashua and Ullman. "Structural Saliency: The Detection of Globally Salient Structures Using a Locally Connected Network." IEEE/CVF International Conference on Computer Vision, 1988. doi:10.1109/CCV.1988.590008

Markdown

[Shashua and Ullman. "Structural Saliency: The Detection of Globally Salient Structures Using a Locally Connected Network." IEEE/CVF International Conference on Computer Vision, 1988.](https://mlanthology.org/iccv/1988/shashua1988iccv-structural/) doi:10.1109/CCV.1988.590008

BibTeX

@inproceedings{shashua1988iccv-structural,
  title     = {{Structural Saliency: The Detection of Globally Salient Structures Using a Locally Connected Network}},
  author    = {Shashua, Amnon and Ullman, Shimon},
  booktitle = {IEEE/CVF International Conference on Computer Vision},
  year      = {1988},
  pages     = {321-327},
  doi       = {10.1109/CCV.1988.590008},
  url       = {https://mlanthology.org/iccv/1988/shashua1988iccv-structural/}
}