CGOF++: Controllable 3D Face Synthesis with Conditional Generative Occupancy Fields
This addresses the problem of inconsistent 3D control in face synthesis for computer vision/graphics applications, representing an incremental improvement over existing 3D-aware methods.
The paper tackles inconsistent face generation under large expression/pose changes in 2D methods by proposing CGOF++, a NeRF-based 3D face synthesis framework that enforces explicit 3D conditions from face priors, achieving more precise 3D controllability than state-of-the-art 2D methods.
Capitalizing on the recent advances in image generation models, existing controllable face image synthesis methods are able to generate high-fidelity images with some levels of controllability, e.g., controlling the shapes, expressions, textures, and poses of the generated face images. However, previous methods focus on controllable 2D image generative models, which are prone to producing inconsistent face images under large expression and pose changes. In this paper, we propose a new NeRF-based conditional 3D face synthesis framework, which enables 3D controllability over the generated face images by imposing explicit 3D conditions from 3D face priors. At its core is a conditional Generative Occupancy Field (cGOF++) that effectively enforces the shape of the generated face to conform to a given 3D Morphable Model (3DMM) mesh, built on top of EG3D [1], a recent tri-plane-based generative model. To achieve accurate control over fine-grained 3D face shapes of the synthesized images, we additionally incorporate a 3D landmark loss as well as a volume warping loss into our synthesis framework. Experiments validate the effectiveness of the proposed method, which is able to generate high-fidelity face images and shows more precise 3D controllability than state-of-the-art 2D-based controllable face synthesis methods.