S^3-Face: SSS-Compliant Facial Reflectance Estimation via Diffusion Priors

Xingyu Ren; Jiankang Deng; Yuhao Cheng; Wenhan Zhu; Yichao Yan; Xiaokang Yang; Stefanos Zafeiriou; Chao Ma

doi:10.1109/cvpr52734.2025.01496

CVPR2025顶会

S^3-Face: SSS-Compliant Facial Reflectance Estimation via Diffusion Priors

Xingyu Ren, Jiankang Deng, Yuhao Cheng, Wenhan Zhu, Yichao Yan, Xiaokang Yang, Stefanos Zafeiriou, Chao Ma

DOI 出版方

2025年份

2顶会引用

摘要

Figure 1 . We introduce S 3 -Face, a comprehensive high-quality facial reflectance estimation method. Our approach leverages a pretrained diffusion model as a reflectance prior, robustly generating subsurface scattering (SSS)-compliant facial reflectance, particularly for hemoglobin and melanin maps. S 3 -Face adeptly handles diverse input subjects and formats, delivering photorealistic rendering results.