IConFace: Identity-Structure Asymmetric Conditioning for Unified Reference-Aware Face Restoration
2026-05-04 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors address the challenge of restoring blurry or damaged face images, especially when important identity details are missing. They introduce IConFace, a system that can use reference images of the same person to help restore the face but also works well without any reference. Their method treats identity information from references and the face structure from the degraded image differently to avoid copying mismatched details. This unified approach improves the clarity and identity accuracy of restored faces whether or not reference images are available.
blind face restorationidentity preservationreference-based restorationno-reference restorationAdaFacecross-attentionlow-rank residualsfacial image degradationidentity anchorspatial structure anchor
Authors
Axi Niu, Jinyang Zhang, Senyan Qing
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.