The real-world adoption of portrait relighting is hindered by dataset domain gaps, camera sensitivity, and computational costs. We address these challenges with Hybrid Domain Knowledge Fusion, a paradigm that fuses the specialized strengths of synthetic, One-Light-at-A-Time (OLAT), and real-world datasets into a compact model. Our approach features specialized prior models hardened by domain-aware adaptation, followed by augmented knowledge distillation into a lightweight student model with multi-domain expertise. Our method demonstrates a 6x to 240x inference speedup while maintaining state-of-the-art (SOTA) visual quality in the experiments. Additionally, we construct a massive, high-fidelity synthetic dataset with diverse ground-truth intrinsics to support our training pipeline.

翻译：人像重光照技术在真实场景中的落地应用受到数据集领域差距、相机灵敏度以及计算成本的制约。针对这些挑战，本文提出了混合领域知识融合范式，该范式将合成数据集、逐光源采集数据集以及真实场景数据集的特定领域优势整合至紧凑模型中。具体而言，本文首先通过领域感知自适应技术构建具有鲁棒性的专用先验模型，随后利用增强型知识蒸馏方法，将多领域专业知识迁移至轻量级学生模型。实验表明，本方法在保持当前最优视觉质量的同时，实现了6倍至240倍的推理加速。此外，为支撑训练流程，本文构建了包含多样化真实本征信息的大规模高保真合成数据集。