UV Volumes for Real-time Rendering of Editable Free-view Human Performance

TL;DR

This paper introduces UV Volumes, a real-time neural volume rendering method for human performance that separates appearance details into 2D neural textures, enabling efficient, editable, and photo-realistic free-view rendering.

Contribution

The paper proposes UV Volumes, a novel approach that encodes high-frequency human appearance into 2D neural textures, improving real-time rendering and editability over prior methods.

Findings

Achieves 30FPS rendering of 960x540 images with photo-realism.

Enables editable free-view human performance with better generalization.

Supports applications like retexturing through neural texture stacks.

Abstract

Neural volume rendering enables photo-realistic renderings of a human performer in free-view, a critical task in immersive VR/AR applications. But the practice is severely limited by high computational costs in the rendering process. To solve this problem, we propose the UV Volumes, a new approach that can render an editable free-view video of a human performer in real-time. It separates the high-frequency (i.e., non-smooth) human appearance from the 3D volume, and encodes them into 2D neural texture stacks (NTS). The smooth UV volumes allow much smaller and shallower neural networks to obtain densities and texture coordinates in 3D while capturing detailed appearance in 2D NTS. For editability, the mapping between the parameterized human model and the smooth texture coordinates allows us a better generalization on novel poses and shapes. Furthermore, the use of NTS enables interesting applications, e.g., retexturing. Extensive experiments on CMU Panoptic, ZJU Mocap, and H36M datasets show that our model can render 960 x 540 images in 30FPS on average with comparable photo-realism to state-of-the-art methods. The project and supplementary materials are available at https://fanegg.github.io/UV-Volumes.