MI-NeRF: Learning a Single Face NeRF from Multiple Identities

arXiv:2403.19920v1 Announce Type: new
Abstract: In this work, we introduce a method that learns a single dynamic neural radiance field (NeRF) from monocular talking face videos of multiple identities. NeRFs have shown remarkable results in modeling the 4D dynamics and appearance of human faces. However, they require per-identity optimization. Although recent approaches have proposed techniques to reduce the training and rendering time, increasing the number of identities can be expensive. We introduce MI-NeRF (multi-identity NeRF), a single unified network that …

abstract arxiv cs.cv dynamic dynamics face however human identity modeling multiple nerf neural radiance field optimization per results type videos work