NDF: Neural Deformable Fields for Dynamic Human Modelling
This work addresses the challenge of accurately modeling dynamic humans for applications like virtual reality or animation, representing an incremental improvement over existing methods.
The authors tackled the problem of dynamic human digitization from multi-view video by proposing Neural Deformable Fields (NDF), which outperforms recent human synthesis methods in synthesizing novel views and poses with detailed appearance.
We propose Neural Deformable Fields (NDF), a new representation for dynamic human digitization from a multi-view video. Recent works proposed to represent a dynamic human body with shared canonical neural radiance fields which links to the observation space with deformation fields estimations. However, the learned canonical representation is static and the current design of the deformation fields is not able to represent large movements or detailed geometry changes. In this paper, we propose to learn a neural deformable field wrapped around a fitted parametric body model to represent the dynamic human. The NDF is spatially aligned by the underlying reference surface. A neural network is then learned to map pose to the dynamics of NDF. The proposed NDF representation can synthesize the digitized performer with novel views and novel poses with a detailed and reasonable dynamic appearance. Experiments show that our method significantly outperforms recent human synthesis methods.