Styleverse: Towards Identity Stylization across Heterogeneous Domains

TL;DR

This paper introduces Styleverse, a novel framework for identity stylization across diverse face styles, leveraging a new dataset FS13, and demonstrates superior fidelity in stylized faces compared to existing methods.

Contribution

The paper presents the first heterogeneous-domain framework for identity stylization, utilizing a single generator and a new dataset FS13 for diverse face styles.

Findings

Styleverse outperforms state-of-the-art methods in identity stylization fidelity.

Introduces a new dataset FS13 with 13 face styles for heterogeneous domain research.

Establishes a quantitative benchmark and qualitative matrix for IDS evaluation.

Abstract

We propose a new challenging task namely IDentity Stylization (IDS) across heterogeneous domains. IDS focuses on stylizing the content identity, rather than completely swapping it using the reference identity. We use an effective heterogeneous-network-based framework $Styleverse$ that uses a single domain-aware generator to exploit the Metaverse of diverse heterogeneous faces, based on the proposed dataset FS13 with limited data. FS13 means 13 kinds of Face Styles considering diverse lighting conditions, art representations and life dimensions. Previous similar tasks, \eg, image style transfer can handle textural style transfer based on a reference image. This task usually ignores the high structure-aware facial area and high-fidelity preservation of the content. However, Styleverse intends to controllably create topology-aware faces in the Parallel Style Universe, where the source facial identity is adaptively styled via AdaIN guided by the domain-aware and reference-aware style embeddings from heterogeneous pretrained models. We first establish the IDS quantitative benchmark as well as the qualitative Styleverse matrix. Extensive experiments demonstrate that Styleverse achieves higher-fidelity identity stylization compared with other state-of-the-art methods.