Simple vs complex temporal recurrences for video saliency prediction
This work addresses video saliency prediction for computer vision applications, but it is incremental as it builds on an existing architecture.
The paper tackled video saliency prediction by modifying an existing neural network with two temporal recurrences, achieving state-of-the-art results on the DHF1K dataset.
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. Source code is available at https://git.io/fjPiB.