This project uses CNN back-propagation to do image saliency detection, and achieved state-of-the-art performance with fast speed.
On Pascal VOC 2012 this method achieves 0.685 F_0.3 score on the test set.
If you hope to use the codes, please cite (the paper has been accepted by IJCNN 2017):
@inproceedings{pan2017fast,
title={A fast method for saliency detection by back-propagating a convolutional neural network and clamping its partial outputs},
author={Pan, Hengyue and Jiang, Hui},
booktitle={Neural Networks (IJCNN), 2017 International Joint Conference on},
pages={3585--3592},
year={2017},
organization={IEEE}
}
Based on MatConvNet
MatConvNet is a MATLAB toolbox implementing Convolutional Neural Networks (CNNs) for computer vision applications. It is simple, efficient, and can run and learn state-of-the-art CNNs. Several example CNNs are included to classify and encode images. Please visit the homepage to know more.