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 text recognition. In this paper we propose an improvement over the original Text Proposals algorithm of Gomez and Karatzas (2016), 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.
@article{arxiv.1702.05089,
title = {Improving Text Proposals for Scene Images with Fully Convolutional Networks},
author = {Dena Bazazian and Raul Gomez and Anguelos Nicolaou and Lluis Gomez and Dimosthenis Karatzas and Andrew D. Bagdanov},
journal= {arXiv preprint arXiv:1702.05089},
year = {2017}
}
Comments
6 pages, 8 figures, International Conference on Pattern Recognition (ICPR) - DLPR (Deep Learning for Pattern Recognition) workshop