FaceFolds: Meshed Radiance Manifolds for Efficient Volumetric Rendering of Dynamic Faces
This work addresses the problem of real-time, high-quality 3D face rendering for applications like gaming and virtual reality, offering an incremental improvement in efficiency and compatibility.
The paper tackles the challenge of efficiently rendering dynamic facial performances with high photorealism by introducing a novel representation that uses meshed radiance manifolds, achieving comparable quality to state-of-the-art neural rendering at significantly higher frame rates.
3D rendering of dynamic face captures is a challenging problem, and it demands improvements on several fronts$\unicode{x2014}$photorealism, efficiency, compatibility, and configurability. We present a novel representation that enables high-quality volumetric rendering of an actor's dynamic facial performances with minimal compute and memory footprint. It runs natively on commodity graphics soft- and hardware, and allows for a graceful trade-off between quality and efficiency. Our method utilizes recent advances in neural rendering, particularly learning discrete radiance manifolds to sparsely sample the scene to model volumetric effects. We achieve efficient modeling by learning a single set of manifolds for the entire dynamic sequence, while implicitly modeling appearance changes as temporal canonical texture. We export a single layered mesh and view-independent RGBA texture video that is compatible with legacy graphics renderers without additional ML integration. We demonstrate our method by rendering dynamic face captures of real actors in a game engine, at comparable photorealism to state-of-the-art neural rendering techniques at previously unseen frame rates.