Abstract

We present a new feature representation method for scene text recognition problem, particularly focusing on improving scene character recognition. Many existing methods rely on Histogram of Oriented Gradient (HOG) or partbased models, which do not span the feature space well for characters in natural scene images, especially given large variation in fonts with cluttered backgrounds. In this work, we propose a discriminative feature pooling method that automatically learns the most informative sub-regions of each scene character within a multi-class classifification framework, whereas each sub-region seamlessly integrates a set of low-level image features through integral images. The proposed feature representation is compact, computationally effificient, and able to effectively model distinctive spatial structures of each individual character class. Extensive experiments conducted on challenging datasets (Chars74K, ICDAR’03, ICDAR’11, SVT) show that our method signifificantly outperforms existing methods on scene character classifification and scene text recognition tasks