Stroke Patches: Customizable Artistic Image Styling Using Regression

TL;DR

This paper introduces a regression-based technique for artistic image styling that uses procedurally generated stroke patches and a U-Net model to produce customizable, stylized images with explicit control over stroke details.

Contribution

It presents a novel, explicit control method for artistic image styling using stroke patches and a regression model, differing from neural style transfer approaches.

Findings

Allows detailed control over stroke composition and style

Produces diverse, customizable artistic renderings

Outperforms some existing style transfer methods in flexibility

Abstract

We present a novel, regression-based method for artistically styling images. Unlike recent neural style transfer or diffusion-based approaches, our method allows for explicit control over the stroke composition and level of detail in the rendered image through the use of an extensible set of stroke patches. The stroke patch sets are procedurally generated by small programs that control the shape, size, orientation, density, color, and noise level of the strokes in the individual patches. Once trained on a set of stroke patches, a U-Net based regression model can render any input image in a variety of distinct, evocative and customizable styles.