End-to-end Handwritten Paragraph Text Recognition Using a Vertical Attention Network
This work provides an improved end-to-end solution for handwritten paragraph text recognition, which is beneficial for document digitization and archival for various industries.
This paper addresses unconstrained handwritten paragraph text recognition by proposing a unified end-to-end model that iteratively processes paragraph images line by line. The model achieves state-of-the-art character error rates of 1.91% on RIMES, 4.45% on IAM, and 3.59% on READ 2016 datasets.
Unconstrained handwritten text recognition remains challenging for computer vision systems. Paragraph text recognition is traditionally achieved by two models: the first one for line segmentation and the second one for text line recognition. We propose a unified end-to-end model using hybrid attention to tackle this task. This model is designed to iteratively process a paragraph image line by line. It can be split into three modules. An encoder generates feature maps from the whole paragraph image. Then, an attention module recurrently generates a vertical weighted mask enabling to focus on the current text line features. This way, it performs a kind of implicit line segmentation. For each text line features, a decoder module recognizes the character sequence associated, leading to the recognition of a whole paragraph. We achieve state-of-the-art character error rate at paragraph level on three popular datasets: 1.91% for RIMES, 4.45% for IAM and 3.59% for READ 2016. Our code and trained model weights are available at https://github.com/FactoDeepLearning/VerticalAttentionOCR.