GaussianSSC: Triplane-Guided Directional Gaussian Fields for 3D Semantic Completion

TL;DR

GaussianSSC introduces a novel two-stage approach for 3D semantic scene completion that leverages Gaussian fields and triplane features to improve occupancy and semantic accuracy efficiently.

Contribution

The paper proposes GaussianSSC, a new method combining Gaussian anchoring and Gaussian-triplane refinement for enhanced 3D scene completion.

Findings

Improves Stage 1 occupancy by +1.0% Recall and +2.0% Precision.

Enhances Stage 2 semantic prediction by +1.8% IoU.

Achieves state-of-the-art results on SemanticKITTI dataset.

Abstract

We present \emph{GaussianSSC}, a two-stage, grid-native and triplane-guided approach to semantic scene completion (SSC) that injects the benefits of Gaussians without replacing the voxel grid or maintaining a separate Gaussian set. We introduce \emph{Gaussian Anchoring}, a sub-pixel, Gaussian-weighted image aggregation over fused FPN features that tightens voxel--image alignment and improves monocular occupancy estimation. We further convert point-like voxel features into a learned per-voxel Gaussian field and refine triplane features via a triplane-aligned \emph{Gaussian--Triplane Refinement} module that combines \emph{local gathering} (target-centric) and \emph{global aggregation} (source-centric). This directional, anisotropic support captures surface tangency, scale, and occlusion-aware asymmetry while preserving the efficiency of triplane representations. On SemanticKITTI~\cite{behley2019semantickitti}, GaussianSSC improves Stage~1 occupancy by +1.0\% Recall, +2.0\% Precision, and +1.8\% IoU over state-of-the-art baselines, and improves Stage~2 semantic prediction by +1.8\% IoU and +0.8\% mIoU.