Gaussian Control with Hierarchical Semantic Graphs in 3D Human Recovery

Hongsheng Wang1,2, Weiyue Zhang2, Sihao Liu2, Xinrui Zhou2, Jing Li2,
Zhanyun Tang2, Shengyu Zhang†1, Fei Wu1 and Feng Lin2


1 Zhejiang University, China      2 Zhejiang Lab, China
Teaser Image

Human Gaussian Control with Hierarchical Semantic Graphs. (a) is a human Gaussian point cloud with semantic labels. (b) is the rendering output of (a). (c) is the result of our method compared with other methods on the Monocap dataset. LPIPS* = LPIPS × 1000.

Abstract

Although 3D Gaussian Splatting (3DGS) has recently made progress in 3D human reconstruction, it primarily relies on 2D pixel-level supervision, overlooking the geometric complexity and topological relationships of different body parts. To address this gap, we introduce the Hierarchical Graph Human Gaussian Control (HUGS) framework for achieving high-fidelity 3D human reconstruction. Our approach involves leveraging explicitly semantic priors of body parts to ensure the consistency of geometric topology, thereby enabling the capture of the complex geometrical and topological associations among body parts. Additionally, we disentangle high-frequency features to refine surface details in body parts. Extensive experiments demonstrate that our method exhibits superior performance in human body reconstruction, particularly in enhancing surface details and accurately reconstructing body part junctions.

Pipeline

Pipeline Image

HUGS framework. We introduce Human Gaussian Control with Hierarchical Semantic Graphs as a method for generating Gaussian humans, ensuring both realistic human appearance and anatomical structure.

Training View Results

GT
Ours
InstantNVR