TIDI-GS: Floater Suppression in 3D Gaussian Splatting for Enhanced Indoor Scene Fidelity

TL;DR

TIDI-GS is a lightweight training framework that effectively removes floaters in 3D Gaussian Splatting, significantly improving scene fidelity and geometric accuracy without major architectural changes.

Contribution

We introduce TIDI-GS, a novel floater pruning method integrated into 3D Gaussian Splatting that enhances scene quality by eliminating artifacts while preserving details.

Findings

Reduces floaters and improves scene fidelity.

Enhances geometric accuracy of 3D reconstructions.

Maintains high-frequency details during cleanup.

Abstract

3D Gaussian Splatting (3DGS) is a technique to create high-quality, real-time 3D scenes from images. This method often produces visual artifacts known as floaters--nearly transparent, disconnected elements that drift in space away from the actual surface. This geometric inaccuracy undermines the reliability of these models for practical applications, which is critical. To address this issue, we introduce TIDI-GS, a new training framework designed to eliminate these floaters. A key benefit of our approach is that it functions as a lightweight plugin for the standard 3DGS pipeline, requiring no major architectural changes and adding minimal overhead to the training process. The core of our method is a floater pruning algorithm--TIDI--that identifies and removes floaters based on several criteria: their consistency across multiple viewpoints, their spatial relationship to other elements, and an importance score learned during training. The framework includes a mechanism to preserve fine details, ensuring that important high-frequency elements are not mistakenly removed. This targeted cleanup is supported by a monocular depth-based loss function that helps improve the overall geometric structure of the scene. Our experiments demonstrate that TIDI-GS improves both the perceptual quality and geometric integrity of reconstructions, transforming them into robust digital assets, suitable for high-fidelity applications.