This 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.
This item is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 3.0 License. View License
Funders:
Science Foundation Ireland (SFI) under grant number SFI/15/SIRG/3283 and SFI/12/RC/2289., framework of project TEC2016- 75976-R, funded by the Spanish Ministerio de Economia y Competitividad., European Regional Development Fund (ERDF)
ID Code:
23543
Deposited On:
10 Jul 2019 11:52 by
Panagiotis Linardos
. Last Modified 27 Oct 2021 12:59
Available Versions of this Item
Temporal recurrences for video saliency prediction. (deposited 10 Jul 2019 11:52)[Currently Displayed]