AvatarMAV: Fast 3D Head Avatar Reconstruction Using Motion-Aware Neural Voxels
This addresses the efficiency bottleneck for researchers and practitioners in facial reenactment, though it is incremental as it builds on existing NeRF and 3DMM frameworks.
The paper tackles the slow training of NeRF-based 3D head avatar reconstruction by proposing AvatarMAV, which uses motion-aware neural voxels to model appearance and expression motion, achieving photo-realistic results in just 5 minutes.
With NeRF widely used for facial reenactment, recent methods can recover photo-realistic 3D head avatar from just a monocular video. Unfortunately, the training process of the NeRF-based methods is quite time-consuming, as MLP used in the NeRF-based methods is inefficient and requires too many iterations to converge. To overcome this problem, we propose AvatarMAV, a fast 3D head avatar reconstruction method using Motion-Aware Neural Voxels. AvatarMAV is the first to model both the canonical appearance and the decoupled expression motion by neural voxels for head avatar. In particular, the motion-aware neural voxels is generated from the weighted concatenation of multiple 4D tensors. The 4D tensors semantically correspond one-to-one with 3DMM expression basis and share the same weights as 3DMM expression coefficients. Benefiting from our novel representation, the proposed AvatarMAV can recover photo-realistic head avatars in just 5 minutes (implemented with pure PyTorch), which is significantly faster than the state-of-the-art facial reenactment methods. Project page: https://www.liuyebin.com/avatarmav.