2015/083 - End-to-End Saliency Mapping via Probability Distribution Prediction
This work concerned with visual attention prediction, specifically predicting eye-fixation maps in still images.
Visual attention has been traditionally used in computer vision domains as a pre-processing step in order to focus later processing on regions of interest in images.
This stage is evermore important as vision models and datasets increase in size. We proposal an end-to-end model which, given an input image, outputs a topographic saliency map. The map is formulated as a generalized Bernoulli distribution, and the model is trained by optimizing a loss function suitable for measuring distances between distributions.