Attention networks for image-to-text.
Jason Poulos, Rafael Valle · arXiv (Cornell University) · 2017
The paper approaches the problem of image-to-text with attention-based encoder-decoder networks that are trained to handle sequences of characters rather than words. We experiment on lines of text from a popular handwriting database with different attention mechanisms for the decoder. The model trained with softmax attention achieves the lowest test error, outperforming several other RNN-based models. Our results show that softmax attention is able to learn a linear alignment whereas the alignment generated by sigmoid attention is linear but much less precise.