%0 Conference Proceedings
%T Improving Text Proposals for Scene Images with Fully Convolutional Networks
%A Dena Bazazian
%A Raul Gomez
%A Anguelos Nicolaou
%A Lluis Gomez
%A Dimosthenis Karatzas
%A Andrew Bagdanov
%B 23rd International Conference on Pattern Recognition Workshops
%D 2016
%F Dena Bazazian2016
%O DAG; LAMP; 600.084
%O exported from refbase (http://refbase.cvc.uab.es/show.php?record=2823), last updated on Mon, 21 Jan 2019 14:14:24 +0100
%X Text Proposals have emerged as a class-dependent version of object proposals – efficient approaches to reduce the search space of possible text object locations in an image. Combined with strong word classifiers, text proposals currently yield top state of the art results in end-to-end scene textrecognition. In this paper we propose an improvement over the original Text Proposals algorithm of [1], combining it with Fully Convolutional Networks to improve the ranking of proposals. Results on the ICDAR RRC and the COCO-text datasets show superior performance over current state-of-the-art.
%U http://refbase.cvc.uab.es/files/BGN2016.pdf