NLDF: Neural Light Dynamic Fields for 3D Talking Head Generation
摘要
Talking head generation based on the neural radiation fields (NeRF) model has shown promising visual effects. However, the slow rendering speed of NeRF seriously limits its application. In this work, a novel Neural Light Dynamic Fields (NLDF) model is proposed aiming to achieve generating high quality 3D talking face with significant speedup. The NLDF represents light fields based on light segments, and a deep network is used to learn the entire light beam’s information at once. In learning the knowledge distillation is applied and the NeRF based synthesized result is used to guide the correct coloration of light segments in NLDF. The propose method effectively represents the facial light dynamics in 3D talking video generation, and it achieves approximately 30 times faster speed compared to state-of-the-art NeRF based method, with comparable generation visual quality.