Paper: http://ceur-ws.org/Vol-2670/MediaEval_19_paper_48.pdf Youtube: https://www.youtube.com/watch?v=bnp0arMsdYc Pierrick Bruneau and Thomas Tamisier, Transfer Learning and Mixed Input Deep Neural Networks for Estimating Flood Severity in News Content. Proc. of MediaEval 2019, 27-29 October 2019, Sophia Antipolis, France. Abstract: This paper describes deep learning approaches which use textual and visual features for flood severity detection in news content. In the context of the MediaEval 2019 Multimedia Satellite task, we test the value of transferring models pre-trained on large related corpora, as well as the improvement brought by dual branch models that combine embeddings output from mixed textual and visual inputs. Presented by Pierrick Bruneau