Paper · arXiv 2312.15258
Human101: Training 100+FPS Human Gaussians in 100s from 1 View
Mingwei Li, Jiachen Tao, Zongxin Yang, Yi Yang
Human101 is a framework that reconstructs high-fidelity dynamic 3D humans from single-view videos in real-time using 3D Gaussian Splatting and Human-centric Forward Gaussian Animation, achieving high rendering speeds and quality.
Abstract
Reconstructing the human body from single-view videos plays a pivotal role in the virtual reality domain. One prevalent application scenario necessitates the rapid reconstruction of high-fidelity 3D digital humans while simultaneously ensuring real-time rendering and interaction. Existing methods often struggle to fulfill both requirements. In this paper, we introduce Human101, a novel framework adept at producing high-fidelity dynamic 3D human reconstructions from 1-view videos by training 3D Gaussians in 100 seconds and rendering in 100+ FPS. Our method leverages the strengths of 3D Gaussian Splatting, which provides an explicit and efficient representation of 3D humans. Standing apart from prior NeRF-based pipelines, Human101 ingeniously applies a Human-centric Forward Gaussian Animation method to deform the parameters of 3D Gaussians, thereby enhancing rendering speed (i.e., rendering 1024-resolution images at an impressive 60+ FPS and rendering 512-resolution images at 100+ FPS). Experimental results indicate that our approach substantially eclipses current methods, clocking up to a 10 times surge in frames per second and delivering comparable or superior rendering quality. Code and demos will be released at https://github.com/longxiang-ai/Human101.