Learning to Sketch with Deep Q Networks and Demonstrated Strokes

TL;DR

This paper introduces Doodle-SDQ, a two-stage deep reinforcement learning system that enables machines to learn doodling by mimicking human strokes and refining their drawing skills without direct supervision.

Contribution

It presents a novel two-stage framework combining supervised stroke imitation and deep Q-learning for autonomous doodling in various media types.

Findings

Doodle-SDQ can learn doodling without step-by-step action supervision.

Pretraining with stroke demonstration enhances performance.

Effective in sketch and watercolor media.

Abstract

Doodling is a useful and common intelligent skill that people can learn and master. In this work, we propose a two-stage learning framework to teach a machine to doodle in a simulated painting environment via Stroke Demonstration and deep Q-learning (SDQ). The developed system, Doodle-SDQ, generates a sequence of pen actions to reproduce a reference drawing and mimics the behavior of human painters. In the first stage, it learns to draw simple strokes by imitating in supervised fashion from a set of strokeaction pairs collected from artist paintings. In the second stage, it is challenged to draw real and more complex doodles without ground truth actions; thus, it is trained with Qlearning. Our experiments confirm that (1) doodling can be learned without direct stepby- step action supervision and (2) pretraining with stroke demonstration via supervised learning is important to improve performance. We further show that Doodle-SDQ is effective at producing plausible drawings in different media types, including sketch and watercolor.