keywords saliency maps; object recognition; egocentric vision; vision modelling; image processing; video processing