HarmoGS: Robust 3D Gaussian Splatting in the Wild via Conflict-Aware Gradient Harmonization
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911687483850752 |
|---|---|
| author | Kang, Yulei Zhu, Tianze Hu, Jian-Fang Lai, Jianhuang Zheng, Wei-Shi |
| author_facet | Kang, Yulei Zhu, Tianze Hu, Jian-Fang Lai, Jianhuang Zheng, Wei-Shi |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_13073 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HarmoGS: Robust 3D Gaussian Splatting in the Wild via Conflict-Aware Gradient Harmonization Kang, Yulei Zhu, Tianze Hu, Jian-Fang Lai, Jianhuang Zheng, Wei-Shi Computer Vision and Pattern Recognition 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. |
| title | HarmoGS: Robust 3D Gaussian Splatting in the Wild via Conflict-Aware Gradient Harmonization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.13073 |