DSS: Synthesizing long Digital Ink using Data augmentation, Style encoding and Split generation
This solves the specific problem of generating realistic long handwriting sequences for digital ink applications, representing a domain-specific incremental improvement.
The paper tackles the problem of synthesizing long-form digital ink text by addressing generalization failures of existing models, achieving a 50% reduction in character error rate compared to baseline RNN and 16% improvement over previous approaches.
As text generative models can give increasingly long answers, we tackle the problem of synthesizing long text in digital ink. We show that the commonly used models for this task fail to generalize to long-form data and how this problem can be solved by augmenting the training data, changing the model architecture and the inference procedure. These methods use contrastive learning technique and are tailored specifically for the handwriting domain. They can be applied to any encoder-decoder model that works with digital ink. We demonstrate that our method reduces the character error rate on long-form English data by half compared to baseline RNN and by 16% compared to the previous approach that aims at addressing the same problem. We show that all three parts of the method improve recognizability of generated inks. In addition, we evaluate synthesized data in a human study and find that people perceive most of generated data as real.