Calligraphic Stylisation Learning with a Physiologically Plausible Model of Movement and Recurrent Neural Networks

TL;DR

This paper introduces a computational framework that learns calligraphic and graffiti stylisation patterns from limited examples, generating new trajectories with similar stylistic qualities using a physiologically plausible movement model and recurrent neural networks.

Contribution

It combines a Recurrent Mixture Density Network with the Sigma Lognormal movement model to learn and generate stylised trajectories from small datasets.

Findings

Successfully learns stylistic patterns from few examples

Generates trajectories that mimic human handwriting and graffiti styles

Uses physiologically plausible movement modeling for realistic results

Abstract

We propose a computational framework to learn stylisation patterns from example drawings or writings, and then generate new trajectories that possess similar stylistic qualities. We particularly focus on the generation and stylisation of trajectories that are similar to the ones that can be seen in calligraphy and graffiti art. Our system is able to extract and learn dynamic and visual qualities from a small number of user defined examples which can be recorded with a digitiser device, such as a tablet, mouse or motion capture sensors. Our system is then able to transform new user drawn traces to be kinematically and stylistically similar to the training examples. We implement the system using a Recurrent Mixture Density Network (RMDN) combined with a representation given by the parameters of the Sigma Lognormal model, a physiologically plausible model of movement that has been shown to closely reproduce the velocity and trace of human handwriting gestures.