IConFace: Identity-Structure Asymmetric Conditioning for Unified Reference-Aware Face Restoration

TL;DR

IConFace is a unified face restoration framework that effectively uses reference images or degrades to improve identity preservation and detail recovery in severely degraded face images.

Contribution

It introduces identity-structure asymmetric conditioning, combining reference and degraded image information for versatile face restoration.

Findings

Improves identity consistency in restored faces.

Enhances fine-detail recovery in degraded images.

Operates effectively with or without reference images.

Abstract

Blind face restoration is highly ill-posed under severe degradation, where identity-critical details may be missing from the degraded input. Same-identity references reduce this ambiguity, but mismatched pose, expression, illumination, age, makeup, or local facial states can lead to overuse of reference appearance. We propose \textbf{IConFace}, a unified reference-aware and no-reference framework with identity--structure asymmetric conditioning. References are distilled into a norm-weighted global AdaFace identity anchor for image-only modulation, while the degraded image is reinforced as the spatial structure anchor through low-rank residuals and block-wise degraded cross-attention with two-route memory. The resulting single checkpoint exploits references when available and falls back to no-reference restoration when absent, improving identity consistency, fine-detail recovery, and degraded-only restoration quality in a unified model.