Multi-Level Net: A Visual Saliency Prediction Model
Abstract
State of the art approaches for saliency prediction are based on Fully Convolutional Networks, in which saliency maps are built using the last layer. In contrast, we here present a novel model that predicts saliency maps exploiting a non-linear combination of features coming from different layers of the network. We also present a new loss function to deal with the imbalance issue on saliency masks. Extensive results on three public datasets demonstrate the robustness of our solution. Our model outperforms the state of the art on SALICON, which is the largest and unconstrained dataset available, and obtains competitive results on MIT300 and CAT2000 benchmarks.
Cite
Text
Cornia et al. "Multi-Level Net: A Visual Saliency Prediction Model." European Conference on Computer Vision Workshops, 2016. doi:10.1007/978-3-319-48881-3_21Markdown
[Cornia et al. "Multi-Level Net: A Visual Saliency Prediction Model." European Conference on Computer Vision Workshops, 2016.](https://mlanthology.org/eccvw/2016/cornia2016eccvw-multilevel/) doi:10.1007/978-3-319-48881-3_21BibTeX
@inproceedings{cornia2016eccvw-multilevel,
title = {{Multi-Level Net: A Visual Saliency Prediction Model}},
author = {Cornia, Marcella and Baraldi, Lorenzo and Serra, Giuseppe and Cucchiara, Rita},
booktitle = {European Conference on Computer Vision Workshops},
year = {2016},
pages = {302-315},
doi = {10.1007/978-3-319-48881-3_21},
url = {https://mlanthology.org/eccvw/2016/cornia2016eccvw-multilevel/}
}