PS-StyleGAN: Illustrative Portrait Sketching using Attention-Based Style Adaptation

TL;DR

PS-StyleGAN introduces an attention-based style transfer method for portrait sketching that enables meaningful edits and style modeling with minimal data and training, outperforming existing techniques.

Contribution

The paper presents Attentive Affine transform blocks integrated into StyleGAN for portrait sketching, allowing style and content edits without fine-tuning the generator.

Findings

Outperforms state-of-the-art portrait sketching methods.

Requires only ~100 paired examples for style modeling.

Enables pose and expression editing without fine-tuning.

Abstract

Portrait sketching involves capturing identity specific attributes of a real face with abstract lines and shades. Unlike photo-realistic images, a good portrait sketch generation method needs selective attention to detail, making the problem challenging. This paper introduces \textbf{Portrait Sketching StyleGAN (PS-StyleGAN)}, a style transfer approach tailored for portrait sketch synthesis. We leverage the semantic $W+$ latent space of StyleGAN to generate portrait sketches, allowing us to make meaningful edits, like pose and expression alterations, without compromising identity. To achieve this, we propose the use of Attentive Affine transform blocks in our architecture, and a training strategy that allows us to change StyleGAN's output without finetuning it. These blocks learn to modify style latent code by paying attention to both content and style latent features, allowing us to adapt the outputs of StyleGAN in an inversion-consistent manner. Our approach uses only a few paired examples ($\sim 100$) to model a style and has a short training time. We demonstrate PS-StyleGAN's superiority over the current state-of-the-art methods on various datasets, qualitatively and quantitatively.