MI-NeRF: Learning a Single Face NeRF from Multiple Identities
This addresses the inefficiency of per-identity optimization in NeRFs for facial modeling, offering a more scalable solution for applications like expression transfer and video synthesis.
The paper tackles the problem of learning a single dynamic neural radiance field (NeRF) from monocular talking face videos of multiple identities, reducing training time and enabling robust synthesis of novel expressions for any input identity.
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 models complex non-rigid facial motion for multiple identities, using only monocular videos of arbitrary length. The core premise in our method is to learn the non-linear interactions between identity and non-identity specific information with a multiplicative module. By training on multiple videos simultaneously, MI-NeRF not only reduces the total training time compared to standard single-identity NeRFs, but also demonstrates robustness in synthesizing novel expressions for any input identity. We present results for both facial expression transfer and talking face video synthesis. Our method can be further personalized for a target identity given only a short video.