HarmoGS: Robust 3D Gaussian Splatting in the Wild via Conflict-Aware Gradient Harmonization

TL;DR

HarmoGS introduces a conflict-aware framework for robust 3D Gaussian Splatting in challenging real-world scenarios, effectively handling distractors and illumination inconsistencies through gradient harmonization and adaptive masking.

Contribution

The paper proposes a novel conflict-aware 3D Gaussian Splatting method that improves robustness by gradient-level optimization and adaptive supervision refinement.

Findings

Achieves state-of-the-art rendering quality on in-the-wild benchmarks.

Effectively suppresses unreliable supervision and resolves view conflicts.

Enhances stability of Gaussian primitives through conflict-aware densification and pruning.

Abstract

In-the-wild 3D Gaussian Splatting remains challenging due to transient distractors and illumination-induced cross-view appearance inconsistencies. Existing methods mainly rely on image-level masking to suppress unreliable supervision, but masking alone cannot fully eliminate residual occlusions or resolve illumination-induced inconsistencies, both of which can introduce conflicting cross-view gradients. These unresolved conflicts may destabilize Gaussian optimization and lead to visible reconstruction artifacts. We propose a conflict-aware 3DGS framework that addresses this problem from both image-space supervision and gradient-level optimization. Semantic Consistency-Guided Masking learns pixel-wise consistency scores to adaptively refine prior masks and suppress unreliable supervision before gradient formation. A dual-view Conflict-Aware Gradient Harmonization strategy further reconciles view-specific gradients by mutually rotating them into an orthogonal configuration, reducing negative directional interference across views. We also introduce conflict-aware densification and pruning to stabilize Gaussian growth and remove persistently conflicting primitives. Extensive experiments on standard in-the-wild benchmarks demonstrate that our method achieves state-of-the-art rendering quality under complex transient distractors and cross-view inconsistencies.