The problem of modeling an animatable 3D human head avatar under lightweight setups is of significant importance but has not been well solved. Existing 3D representations either perform well in the realism of portrait images synthesis or the accuracy of expression control, but not both. To address the problem, we introduce a novel hybrid explicit-implicit 3D representation, Facial Model Conditioned Neural Radiance Field, which integrates the expressiveness of NeRF and the prior information from the parametric template. At the core of our representation, a synthetic-renderings-based condition method is proposed to fuse the prior information from the parametric model into the implicit field without constraining its topological flexibility. Besides, based on the hybrid representation, we properly overcome the inconsistent shape issue presented in existing methods and improve the animation stability. Moreover, by adopting an overall GAN-based architecture using an image-to-image translation network, we achieve high-resolution, realistic and view-consistent synthesis of dynamic head appearance. Experiments demonstrate that our method can achieve state-of-the-art performance for 3D head avatar animation compared with previous methods.

在轻量级设置下建模可动画 3D 人体头部头像的问题非常重要，但尚未得到很好的解决。现有的 3D 表示要么在人像图像合成的真实感方面表现良好，要么在表情控制的准确性方面表现良好，但无法两者兼而有之。为了解决这个问题，我们引入了一种新颖的混合显式-隐式 3D 表示，面部模型条件神经辐射场，它集成了 NeRF 的表达能力和参数模板的先验信息。在我们表示的核心，提出了一种基于合成渲染的条件方法，将参数模型的先验信息融合到隐式场中，而不限制其拓扑灵活性。此外，基于混合表示，我们妥善克服了现有方法中出现的形状不一致问题，提高了动画的稳定性。此外，通过采用基于 GAN 的整体架构，使用图像到图像转换网络，我们实现了动态头部外观的高分辨率、真实性和视图一致的合成。实验表明，与之前的方法相比，我们的方法可以实现 3D 头部头像动画最先进的性能。