Convolutional-deconvolution networks can be adopted to perform end-to-end saliency detection. But, they do not work well with objects of multiple scales. To overcome such a limitation, in this work, we propose a recurrent attentional convolutional-deconvolution network (RACDNN). Using spatial transformer and recurrent network units, RACDNN is able to iteratively attend to selected image sub-regions to perform saliency refinement progressively. Besides tackling the scale problem, RACDNN can also learn context-aware features from past iterations to enhance saliency refinement in future iterations. Experiments on several challenging saliency detection datasets validate the effectiveness of RACDNN, and show that RACDNN outperforms state-of-the-art saliency detection methods.
@article{arxiv.1604.03227,
title = {Recurrent Attentional Networks for Saliency Detection},
author = {Jason Kuen and Zhenhua Wang and Gang Wang},
journal= {arXiv preprint arXiv:1604.03227},
year = {2016}
}