IConFace: Identity-Structure Asymmetric Conditioning for Unified Reference-Aware Face Restoration
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915978288300032 |
|---|---|
| author | Niu, Axi Zhang, Jinyang Qing, Senyan |
| author_facet | Niu, Axi Zhang, Jinyang Qing, Senyan |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_02814 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | IConFace: Identity-Structure Asymmetric Conditioning for Unified Reference-Aware Face Restoration Niu, Axi Zhang, Jinyang Qing, Senyan Computer Vision and Pattern Recognition Artificial Intelligence 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. |
| title | IConFace: Identity-Structure Asymmetric Conditioning for Unified Reference-Aware Face Restoration |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2605.02814 |