Endo-G$^{2}$T: Geometry-Guided & Temporally Aware Time-Embedded 4DGS For Endoscopic Scenes

TL;DR

Endo-G$^{2}$T introduces a geometry-guided, temporally aware training scheme for 4D Gaussian splatting that improves endoscopic scene reconstruction by anchoring geometry early and maintaining temporal consistency.

Contribution

The paper proposes a novel training scheme combining geometry-guided prior distillation, a time-embedded Gaussian field, and keyframe-constrained streaming for improved 4D scene reconstruction.

Findings

Achieves state-of-the-art results on EndoNeRF and StereoMIS-P1 datasets.

Enhances temporal coherence and geometric accuracy in endoscopic scene modeling.

Improves efficiency and stability in long-horizon dynamic scene reconstruction.

Abstract

Endoscopic (endo) video exhibits strong view-dependent effects such as specularities, wet reflections, and occlusions. Pure photometric supervision misaligns with geometry and triggers early geometric drift, where erroneous shapes are reinforced during densification and become hard to correct. We ask how to anchor geometry early for 4D Gaussian splatting (4DGS) while maintaining temporal consistency and efficiency in dynamic endoscopic scenes. Thus, we present Endo-G$^{2}$T, a geometry-guided and temporally aware training scheme for time-embedded 4DGS. First, geo-guided prior distillation converts confidence-gated monocular depth into supervision with scale-invariant depth and depth-gradient losses, using a warm-up-to-cap schedule to inject priors softly and avoid early overfitting. Second, a time-embedded Gaussian field represents dynamics in XYZT with a rotor-like rotation parameterization, yielding temporally coherent geometry with lightweight regularization that favors smooth motion and crisp opacity boundaries. Third, keyframe-constrained streaming improves efficiency and long-horizon stability through keyframe-focused optimization under a max-points budget, while non-keyframes advance with lightweight updates. Across EndoNeRF and StereoMIS-P1 datasets, Endo-G$^{2}$T achieves state-of-the-art results among monocular reconstruction baselines.