Learning to Sketch with Deep Q Networks and Demonstrated Strokes
This work addresses the challenge of automated drawing generation for creative AI applications, but it is incremental as it builds on existing reinforcement learning and imitation learning techniques.
The paper tackled the problem of teaching a machine to doodle by proposing a two-stage learning framework that combines stroke demonstration with deep Q-learning, resulting in a system that generates plausible drawings in various media types without direct step-by-step action supervision.
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.