DisplaceNet: Recognising Displaced People from Images by Exploiting Dominance Level

Every year millions of men, women and children are forced to leave theirhomes and seek refuge from wars, human rights violations, persecution, andnatural disasters. The number of forcibly displaced people came at a recordrate of 44,400 every day throughout 2017, raising the cumulative total to 68.5million at the years end, overtaken the total population of the United Kingdom.Up to 85% of the forcibly displaced find refuge in low- and middle-incomecountries, calling for increased humanitarian assistance worldwide. To reducethe amount of manual labour required for human-rights-related image analysis,we introduce DisplaceNet, a novel model which infers potential displaced peoplefrom images by integrating the control level of the situation and conventionalconvolutional neural network (CNN) classifier into one framework for imageclassification. Experimental results show that DisplaceNet achieves up to 4%coverage-the proportion of a data set for which a classifier is able to producea prediction-gain over the sole use of a CNN classifier. Our dataset, codes andtrained models will be available online athttps://github.com/GKalliatakis/DisplaceNet.