Learning To Paint with RL
This problem aims at training an agent to generate paintings using stroke based sequential actions. We plan to break down the given target image into a set number of strokes, using a RL agent, which can generate a painted image. Compare On Policy (PPO) and Off Policy (DDPG) methods for solving the problem. We also aim at incorporating style transfer while painting.
- Learning to Paint is a complex process and often requires considerable time and effort to master the art
- Making an AI agent learn how to reproduce an existing painting by decomposing it into a sequence of brush strokes across the canvas.
- The training process of such an agent does not require any considerable vast experience.
- The problem statement has multiple applications in real life such as handwriting reproduction, painting restoration etc.
- Additionally using concepts from neural style transfer, the reward for the agent can be adjusted in order to incorporate style elements from a reference image.
For results, and presentation see this
For training and testing on MNIST
For training and testing on celeb
For training Style transfer RL on MNIST
Style Transfer on MNIST
Using DDPG on CelebA
Using PPO on MNIST
Version : 1.0.0
- Asmit Kumar Singh - IIIT-Delhi - Other Work
- Suchet Aggarwal - IIIT-Delhi - Other Work
- Pragya Sethi - IIIT-Delhi - Other Work