MaRO-GS: Mask-Robust Object-Centric Gaussian Splatting from Inconsistent Multi-view Masks

2026-10-05

AI Summary

This study addresses the low reconstruction accuracy and computational redundancy in 3D Gaussian object reconstruction caused by multi-view mask inconsistencies. To this end, we propose a mask-robust, object-level 3D Gaussian Splatting framework that departs from conventional full-scene optimization paradigms by directly optimizing the Gaussian primitives of target objects. The core innovations include a mask-reliability view filtering mechanism to suppress noise interference, an object-support density control strategy to eliminate background redundancy, and a silhouette alignment loss function to enhance geometric consistency. Experimental results demonstrate that the proposed framework significantly improves PSNR, segmentation accuracy, and computational efficiency, achieving a notable PSNR gain of 2.05 dB on small-object datasets.

Abstract

We address the challenge of accurate 3D object reconstruction from multi-view images in Gaussian Splatting. Existing object-level 3DGS methods reconstruct the entire scene rather than directly optimizing the target object, even when only the target object is needed, which incurs substantial computational overhead. They also rely on 2D segmentation masks to associate Gaussians with objects, but these masks are often inconsistent across views. Such inconsistencies corrupt Gaussian optimization and produce incorrectly supervised Gaussians that degrade object reconstruction fidelity. To overcome these limitations, we propose MaRO-GS, a 3DGS framework that directly optimizes target-object Gaussians from object-masked multi-view images and remains robust to inconsistent supervision. For reliable supervision, mask-reliability view filtering excludes unreliable views. Object-supported Gaussian density control suppresses Gaussians irrelevant to the target object and prevents background densification, while Silhouette-aligned Object Loss maintains object-focused optimization. Extensive experiments across diverse datasets demonstrate that MaRO-GS improves PSNR, segmentation accuracy, and computational efficiency, with the largest PSNR gain of 2.05 dB on the small-object LERF-Mask dataset.

Problem

• 3D object reconstruction
• Gaussian Splatting
• multi-view masks inconsistency
• object-centric optimization
• computational overhead

Innovation

• Gaussian Splatting
• Object-Centric Reconstruction
• Mask Inconsistency
• Density Control
• Silhouette-aligned Loss