SalGAN: Visual Saliency Prediction with Generative Adversarial Networks

01/04/2017
by   Junting Pan, et al.
0

We introduce SalGAN, a deep convolutional neural network for visual saliency prediction trained with adversarial examples. The first stage of the network consists of a generator model whose weights are learned by back-propagation computed from a binary cross entropy (BCE) loss over downsampled versions of the saliency maps. The resulting prediction is processed by a discriminator network trained to solve a binary classification task between the saliency maps generated by the generative stage and the ground truth ones. Our experiments show how adversarial training allows reaching state-of-the-art performance across different metrics when combined with a widely-used loss function like BCE.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset