Grangier, DavidBengio, Samy2010-02-112010-02-112010-02-11200610.1007/11840930_3https://infoscience.epfl.ch/handle/20.500.14299/46802This work presents a neural network for the retrieval of images from text queries. The proposed network is composed of two main modules: the first one extracts a global picture representation from local block descriptors while the second one aims at solving the retrieval problem from the extracted representation. Both modules are trained jointly to minimize a loss related to the retrieval performance. This approach is shown to be advantageous when compared to previous models relying on unsupervised feature extraction: average precision over Corel queries reaches 26.2\% for our model, which should be compared to 21.6\% for PAMIR, the best alternative.A Neural Network to Retrieve Images from Text Queriestext::conference output::conference proceedings::conference paper