TSPE-GS: Probabilistic Depth Extraction for Semi-Transparent Surface Reconstruction via 3D Gaussian Splatting

TL;DR

TSPE-GS introduces a probabilistic approach to 3D Gaussian Splatting, enabling accurate reconstruction of semi-transparent surfaces by modeling multi-modal depth and opacity distributions, overcoming previous single-peak limitations.

Contribution

It proposes a novel probabilistic sampling method that models multi-surface transparency and depth, improving semi-transparent surface reconstruction without additional training.

Findings

Significantly improves semi-transparent surface reconstruction accuracy.

Maintains performance on opaque scenes.

Generalizes to other Gaussian-based methods without extra training.

Abstract

3D Gaussian Splatting offers a strong speed-quality trade-off but struggles to reconstruct semi-transparent surfaces because most methods assume a single depth per pixel, which fails when multiple surfaces are visible. We propose TSPE-GS (Transparent Surface Probabilistic Extraction for Gaussian Splatting), which uniformly samples transmittance to model a pixel-wise multi-modal distribution of opacity and depth, replacing the prior single-peak assumption and resolving cross-surface depth ambiguity. By progressively fusing truncated signed distance functions, TSPE-GS reconstructs external and internal surfaces separately within a unified framework. The method generalizes to other Gaussian-based reconstruction pipelines without extra training overhead. Extensive experiments on public and self-collected semi-transparent and opaque datasets show TSPE-GS significantly improves semi-transparent geometry reconstruction while maintaining performance on opaque scenes.