HeroGS: Hierarchical Guidance for Robust 3D Gaussian Splatting under Sparse Views

TL;DR

HeroGS introduces a hierarchical guidance framework that enhances 3D Gaussian Splatting under sparse views, improving reconstruction quality by regularizing distributions and refining details through multi-level supervision.

Contribution

The paper proposes HeroGS, a novel hierarchical guidance framework that converts sparse supervision into dense guidance and refines Gaussian distributions at multiple levels for robust 3D reconstruction.

Findings

Outperforms state-of-the-art methods under sparse-view conditions

Achieves higher structural fidelity and rendering quality

Effectively regularizes Gaussian distributions for better 3D reconstructions

Abstract

3D Gaussian Splatting (3DGS) has recently emerged as a promising approach in novel view synthesis, combining photorealistic rendering with real-time efficiency. However, its success heavily relies on dense camera coverage; under sparse-view conditions, insufficient supervision leads to irregular Gaussian distributions, characterized by globally sparse coverage, blurred background, and distorted high-frequency areas. To address this, we propose HeroGS, Hierarchical Guidance for Robust 3D Gaussian Splatting, a unified framework that establishes hierarchical guidance across the image, feature, and parameter levels. At the image level, sparse supervision is converted into pseudo-dense guidance, globally regularizing the Gaussian distributions and forming a consistent foundation for subsequent optimization. Building upon this, Feature-Adaptive Densification and Pruning (FADP) at the feature level leverages low-level features to refine high-frequency details and adaptively densifies Gaussians in background regions. The optimized distributions then support Co-Pruned Geometry Consistency (CPG) at parameter level, which guides geometric consistency through parameter freezing and co-pruning, effectively removing inconsistent splats. The hierarchical guidance strategy effectively constrains and optimizes the overall Gaussian distributions, thereby enhancing both structural fidelity and rendering quality. Extensive experiments demonstrate that HeroGS achieves high-fidelity reconstructions and consistently surpasses state-of-the-art baselines under sparse-view conditions.