Learning Dynamic Tetrahedra for High-Quality Talking Head Synthesis

arXiv:2402.17364v1 Announce Type: new
Abstract: Recent works in implicit representations, such as Neural Radiance Fields (NeRF), have advanced the generation of realistic and animatable head avatars from video sequences. These implicit methods are still confronted by visual artifacts and jitters, since the lack of explicit geometric constraints poses a fundamental challenge in accurately modeling complex facial deformations. In this paper, we introduce Dynamic Tetrahedra (DynTet), a novel hybrid representation that encodes explicit dynamic meshes by neural networks to ensure geometric …

abstract advanced arxiv avatars challenge constraints cs.cv dynamic fields head nerf neural radiance fields quality synthesis type video visual