The code is for paper "Pattern Mining Saliency" by Yuqiu Kong, Lijun Wang, Xiuping Liu, Huchuan Lu, Xiang Ruan [1].

To appear the 14th European Conference on Computer Vision (ECCV 2016), Amsterdam, Netherlands, 2016.

Email: jinghongkyq@gmail.com

******************************************************************************************************************

The code is tested on Ubuntu 14.04 with MATLAB R2014b.
******************************************************************************************************************
Usage:

1. demo_single.m

Demo for single model promotion. Input an image and its corresponding saliency map generated by existing saliency model (i.e., CA[2]), output a refined saliency map (i.e., CA-PM).


2. demo_joint.m

Demo for joint promotion. Input an image and its corresponding saliency maps generated by several saliency models (i.e., CA, CB[3], LR[4]), output a jointly refined saliency map (i.e., CACBLR-PM)  



Toolbox:

1. Apriori algorithm

We use this toolbox (http://www.borgelt.net/src/apriori.tar.gz) to excute rule mining task.
You need to uncompress it and run 'make' to compile it in the 'apriori/apriori/src'. 
Detailed usage of this package can be found in http://www.borgelt.net/doc/apriori/apriori.html.


2. vlfeat toolbox
The code utilizes the vlfeat-0.9.20-bin external toolbox.
The download link: http://www.vlfeat.org/download/
You need to uncompress it to './PMcode/'.


References:

[1] Kong, Y., Wang, L., Liu, X., Lu, H., Ruan, X.: Pattern mining saliency. In: ECCV. (2016) 

[2] Goferman, S., Zelnik-Manor, L., Tal., A.: Context-aware saliency detection. In: CVPR. (2010)

[3] Jiang, H., Wang, J., Yuan, Z., Liu, T., Zheng, N., Li, S.: Automatic salient object segmentation based on context and shape prior. In: BMVC. (2011)

[4] Shen, X.S., Wu, Yu.: A unified approach to salient object detectioin via low rank matrix recovery. In: CVPR. (2012)