B\'ezierSketch: A generative model for scalable vector sketches

TL;DR

BézierSketch is a new generative model for scalable, high-resolution vector sketches that uses Bézier curves for stroke embedding, enabling better modeling of long sketches compared to previous methods.

Contribution

It introduces a novel inverse graphics approach for stroke embedding into Bézier curves, allowing scalable, high-resolution sketch generation of longer sequences.

Findings

Outperforms previous models on Quick, Draw! benchmark

Produces scalable, high-resolution vector sketches

Effectively models long sketches with improved capacity

Abstract

The study of neural generative models of human sketches is a fascinating contemporary modeling problem due to the links between sketch image generation and the human drawing process. The landmark SketchRNN provided breakthrough by sequentially generating sketches as a sequence of waypoints. However this leads to low-resolution image generation, and failure to model long sketches. In this paper we present B\'ezierSketch, a novel generative model for fully vector sketches that are automatically scalable and high-resolution. To this end, we first introduce a novel inverse graphics approach to stroke embedding that trains an encoder to embed each stroke to its best fit B\'ezier curve. This enables us to treat sketches as short sequences of paramaterized strokes and thus train a recurrent sketch generator with greater capacity for longer sketches, while producing scalable high-resolution results. We report qualitative and quantitative results on the Quick, Draw! benchmark.