Simple vs complex temporal recurrences for video saliency prediction
Panagiotis Linardos (Insight Center for Data Analytics), Eva Mohedano (Insight Center for Data Analytics), Juan Jose Nieto (Insight Center for Data Analytics), Noel O'Connor (Dublin City University (DCU)), Xavier Giro-i-Nieto (Universitat Politecnica de Catalunya), Kevin McGuinness (Insight Centre for Data Analytics) AbstractThis paper investigates modifying an existing neural network architecture for static saliency prediction using two types of recurrences that integrate information from the temporal domain. The first modification is the addition of a ConvLSTM within the architecture, while the second is a conceptually simple exponential moving average of an internal convolutional state. We use weights pre-trained on the SALICON dataset and fine-tune our model on DHF1K. Our results show that both modifications achieve state-of-the-art results and produce similar saliency maps. Source code is available at \url{https://git.io/fjPiB}.
DOI
10.5244/C.33.185
https://dx.doi.org/10.5244/C.33.185
Files
Paper (PDF)BibTeX
@inproceedings{BMVC2019,
title={Simple vs complex temporal recurrences for video saliency prediction},
author={Panagiotis Linardos and Eva Mohedano and Juan Jose Nieto and Noel O'Connor and Xavier Giro-i-Nieto and Kevin McGuinness},
year={2019},
month={September},
pages={185.1--185.12},
articleno={185},
numpages={12},
booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
publisher={BMVA Press},
editor={Kirill Sidorov and Yulia Hicks},
doi={10.5244/C.33.185},
url={https://dx.doi.org/10.5244/C.33.185}
}
title={Simple vs complex temporal recurrences for video saliency prediction},
author={Panagiotis Linardos and Eva Mohedano and Juan Jose Nieto and Noel O'Connor and Xavier Giro-i-Nieto and Kevin McGuinness},
year={2019},
month={September},
pages={185.1--185.12},
articleno={185},
numpages={12},
booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
publisher={BMVA Press},
editor={Kirill Sidorov and Yulia Hicks},
doi={10.5244/C.33.185},
url={https://dx.doi.org/10.5244/C.33.185}
}