English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Convolutional Neural Networks (CNN) have shown promising results for the taskof Handwritten Text Recognition (HTR) but they still fall behind RecurrentNeural Networks (RNNs)/Transformer based models in terms of performance. Inthis paper, we propose a CNN based architecture that bridges this gap. Ourwork, Easter2.0, is composed of multiple layers of 1D Convolution, BatchNormalization, ReLU, Dropout, Dense Residual connection, Squeeze-and-Excitationmodule and make use of Connectionist Temporal Classification (CTC) loss. Inaddition to the Easter2.0 architecture, we propose a simple and effective dataaugmentation technique 'Tiling and Corruption (TACO)' relevant for the task ofHTR/OCR. Our work achieves state-of-the-art results on IAM handwriting databasewhen trained using only publicly available training data. In our experiments,we also present the impact of TACO augmentations and Squeeze-and-Excitation(SE) on text recognition accuracy. We further show that Easter2.0 is suitablefor few-shot learning tasks and outperforms current best methods includingTransformers when trained on limited amount of annotated data. Code and modelis available at: https://github.com/kartikgill/Easter2

Easter2.0: Improving convolutional models for handwritten text recognition