This folder provides reference code for the paper "Visual Saliency Prediction Using a Mixture of Deep Neural Networks".

Before running the code you will need to download the following pre-trained networks into the models folder:

• Weights VGG-16: vgg16_weights.h5
• Weights ML-Net: mlnet_salicon.h5

Install the required libraries using pip:

pip install -r requirements.txt

You will also need to download the CAT2000 dataset

First set the path to the dataset in the param.py file (parameter BASEPATH). Then run the code using:

python train.py

To test the model run

python test.py

We also include the code to fine-tune the ML-net model. This code can be run using the "baseline_train.py" file.