FaceChain-FACT: Face Adapter with Decoupled Training for Identity-preserved Personalization

TL;DR

FaceChain-FACT introduces a novel framework that enhances identity-preserved personalized image generation by decoupling identity features and training processes, improving controllability and fidelity without sacrificing diversity.

Contribution

The paper proposes the Face Adapter with deCoupled Training (FACT), a new method that decouples identity features and training to improve personalized face generation.

Findings

FACT improves controllability and fidelity in face generation.

The method maintains diversity and identity preservation.

Extensive experiments validate the effectiveness of the approach.

Abstract

In the field of human-centric personalized image generation, the adapter-based method obtains the ability to customize and generate portraits by text-to-image training on facial data. This allows for identity-preserved personalization without additional fine-tuning in inference. Although there are improvements in efficiency and fidelity, there is often a significant performance decrease in test following ability, controllability, and diversity of generated faces compared to the base model. In this paper, we analyze that the performance degradation is attributed to the failure to decouple identity features from other attributes during extraction, as well as the failure to decouple the portrait generation training from the overall generation task. To address these issues, we propose the Face Adapter with deCoupled Training (FACT) framework, focusing on both model architecture and training strategy. To decouple identity features from others, we leverage a transformer-based face-export encoder and harness fine-grained identity features. To decouple the portrait generation training, we propose Face Adapting Increment Regularization~(FAIR), which effectively constrains the effect of face adapters on the facial region, preserving the generative ability of the base model. Additionally, we incorporate a face condition drop and shuffle mechanism, combined with curriculum learning, to enhance facial controllability and diversity. As a result, FACT solely learns identity preservation from training data, thereby minimizing the impact on the original text-to-image capabilities of the base model. Extensive experiments show that FACT has both controllability and fidelity in both text-to-image generation and inpainting solutions for portrait generation.