The Cursive Transformer
This addresses the underexplored challenge of handwriting data generation for applications in digital writing or accessibility, though it is incremental as it adapts existing transformer methods to a new domain.
The paper tackled the problem of generating realistic cursive handwriting by introducing a novel tokenization scheme that converts pen stroke offsets to polar coordinates and discretizes them for training a standard GPT model, achieving realistic generation with only 3,500 handwritten words and simple data augmentations.
Transformers trained on tokenized text, audio, and images can generate high-quality autoregressive samples. But handwriting data, represented as sequences of pen coordinates, remains underexplored. We introduce a novel tokenization scheme that converts pen stroke offsets to polar coordinates, discretizes them into bins, and then turns them into sequences of tokens with which to train a standard GPT model. This allows us to capture complex stroke distributions without using any specialized architectures (eg. the mixture density network or the self-advancing ASCII attention head from Graves 2014). With just 3,500 handwritten words and a few simple data augmentations, we are able to train a model that can generate realistic cursive handwriting. Our approach is simpler and more performant than previous RNN-based methods.