saliencyfilters: A Python implementation of the Saliency Filters method

saliencyfilters is a Python implementation of the Saliency Filters method, which is an algorithm to estimate salient region on images.

References

Perazzi, Federico, et al. "Saliency filters: Contrast based filtering for salient region detection." Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on. IEEE, 2012.

Required Packages

• numpy
• scipy
• scikit-image

Installation

pip install -r requirements.txt
python setup.py install

Usage

Sample code is provided in demo.py.

python demo.py sample.jpg output.jpg

Notes

Our implementation does not use Permutohedral Lattice embedding as stated in the paper. Hence, the code requires $O(N^2)$ operations. The results are pretty much similar but a little bit slower in terms of running time.