MEGA: Object-Level Mesh Extraction from 3D Gaussian Splatting via Spatial Visual Distillation

2026-10-01

AI Summary

This study addresses the challenge of extracting object-level watertight meshes from 3D Gaussian Splatting (3DGS) representations, a task where existing methods often fall short. To overcome this limitation, we propose a "segment-then-reconstruct" framework. The core innovation lies in the introduction of spatial visual distillation, which integrates mask-guided neural surface reconstruction with multi-view photometric supervision to achieve high-fidelity object-level geometry recovery in complex scenes. Extensive experiments demonstrate that our method attains state-of-the-art performance across multiple benchmarks, effectively supporting high-quality physical interactions and photorealistic rendering. Ultimately, this work establishes a novel paradigm for object-level mesh extraction from 3D Gaussian Splatting, bridging the gap between radiance field representations and explicit geometric models suitable for downstream applications.

Abstract

Mesh extraction from 3D Gaussian Splatting (3DGS) aims to endow 3D Gaussians with accurate geometric structures, enabling explicit and precise 3D occupancy. However, existing methods primarily focus on scene-level mesh extraction, making them unable to represent object-level occupancy and often resulting in non-watertight surfaces. To overcome these limitations, we propose \textbf{MEGA} (\underline{M}esh \underline{E}xtraction from \underline{GA}ussians), a ``segment-then-mesh'' framework for extracting object-level, watertight meshes from complex 3DGS scenes. At the core of MEGA are \textbf{Spatial Visual Distillation (SVD)} and a mask-guided neural surface reconstruction module. SVD treats the 3DGS model as a teacher, sampling diverse camera poses and rendering the corresponding views of each segmented object. These observations are then used to train a mesh reconstruction model through photometric supervision. Extensive experiments on several widely used benchmarks demonstrate that MEGA achieves state-of-the-art performance in recovering accurate object-level 3D occupancy. Moreover, MEGA enables complex physical interactions by combining high-quality object-level meshes for geometric occupancy with 3DGS representations for photorealistic rendering.

Problem

• 3D Gaussian Splatting
• mesh extraction
• object-level occupancy
• watertight surfaces

Innovation

• 3D Gaussian Splatting
• Spatial Visual Distillation
• Object-Level Mesh Extraction
• Neural Surface Reconstruction
• Watertight Mesh