Optimizing ID Consistency in Multimodal Large Models: Facial Restoration via Alignment, Entanglement, and Disentanglement

TL;DR

This paper introduces EditedID, a novel framework that enhances facial identity consistency in multimodal large models during portrait editing by aligning, disentangling, and selectively entangling features, achieving state-of-the-art results.

Contribution

Proposes a training-free, plug-and-play framework with three key components to improve facial ID preservation in multimodal editing models, addressing cross-source biases and feature contamination.

Findings

Achieves state-of-the-art facial ID consistency

Effective in open-world, multi-person scenarios

Provides a practical, deployable solution

Abstract

Multimodal editing large models have demonstrated powerful editing capabilities across diverse tasks. However, a persistent and long-standing limitation is the decline in facial identity (ID) consistency during realistic portrait editing. Due to the human eye's high sensitivity to facial features, such inconsistency significantly hinders the practical deployment of these models. Current facial ID preservation methods struggle to achieve consistent restoration of both facial identity and edited element IP due to Cross-source Distribution Bias and Cross-source Feature Contamination. To address these issues, we propose EditedID, an Alignment-Disentanglement-Entanglement framework for robust identity-specific facial restoration. By systematically analyzing diffusion trajectories, sampler behaviors, and attention properties, we introduce three key components: 1) Adaptive mixing strategy that aligns cross-source latent representations throughout the diffusion process. 2) Hybrid solver that disentangles source-specific identity attributes and details. 3) Attentional gating mechanism that selectively entangles visual elements. Extensive experiments show that EditedID achieves state-of-the-art performance in preserving original facial ID and edited element IP consistency. As a training-free and plug-and-play solution, it establishes a new benchmark for practical and reliable single/multi-person facial identity restoration in open-world settings, paving the way for the deployment of multimodal editing large models in real-person editing scenarios. The code is available at https://github.com/NDYBSNDY/EditedID.