Few-Shot Generative Model Adaption via Identity Injection and Preservation

TL;DR

This paper introduces I$^2$P, a novel method for few-shot generative model adaptation that preserves source identity knowledge during target domain adaptation, improving image quality with limited data.

Contribution

The paper proposes Identity Injection and Preservation (I$^2$P), a new approach that maintains source identity in few-shot generative model adaptation, addressing mode collapse and forgetting issues.

Findings

Significant improvement over state-of-the-art methods.

Effective identity preservation demonstrated on multiple datasets.

Enhanced image quality with fewer samples.

Abstract

Training generative models with limited data presents severe challenges of mode collapse. A common approach is to adapt a large pretrained generative model upon a target domain with very few samples (fewer than 10), known as few-shot generative model adaptation. However, existing methods often suffer from forgetting source domain identity knowledge during adaptation, which degrades the quality of generated images in the target domain. To address this, we propose Identity Injection and Preservation (I$^2$P), which leverages identity injection and consistency alignment to preserve the source identity knowledge. Specifically, we first introduce an identity injection module that integrates source domain identity knowledge into the target domain's latent space, ensuring the generated images retain key identity knowledge of the source domain. Second, we design an identity substitution module, which includes a style-content decoupler and a reconstruction modulator, to further enhance source domain identity preservation. We enforce identity consistency constraints by aligning features from identity substitution, thereby preserving identity knowledge. Both quantitative and qualitative experiments show that our method achieves substantial improvements over state-of-the-art methods on multiple public datasets and 5 metrics.