Corpus ID: 16734174

Deep Structured Output Learning for Unconstrained Text Recognition

@article{Jaderberg2015DeepSO,
  title={Deep Structured Output Learning for Unconstrained Text Recognition},
  author={Max Jaderberg and K. Simonyan and A. Vedaldi and Andrew Zisserman},
  journal={CoRR},
  year={2015},
  volume={abs/1412.5903}
}

Max Jaderberg, K. Simonyan, +1 author Andrew Zisserman

Published 2015

Computer Science

CoRR

We develop a representation suitable for the unconstrained recognition of words in natural images: the general case of no fixed lexicon and unknown length. To this end we propose a convolutional neural network (CNN) based architecture which incorporates a Conditional Random Field (CRF) graphical model, taking the whole word image as a single input. The unaries of the CRF are provided by a CNN that predicts characters at each position of the output, while higher order terms are provided by… CONTINUE READING

View PDF on arXiv

Save to Library

Create Alert

Cite

Launch Research Feed