Saliency Detection with Side Information on Railway Transportation Dataset

with the development of intelligent railway transportation, accurate location of wayside track assets like traffic signals, mileposts, switches, crossings make essential sense. In this paper, we propose a novel saliency detection model with deep learning and side information aimed at railway transportation dataset. We firstly construct a new railway transportation dataset, which includes 1282 images from the real videos taken by the camera on moving trains and collected from the Internet. Extensive experiments show the effectiveness of our model which can accurately locate the salient object in the dataset.