EndoPrior-GS: Dynamic Endoscopic Reconstruction with a Joint Texture Prior

2026-09-29

AI Summary

This study addresses the severe degradation in 3D Gaussian Splatting (3DGS) rendering quality during dynamic endoscopic reconstruction, which is primarily caused by pseudo-geometry and abrupt illumination variations. To overcome these challenges, this work proposes a joint texture prior that integrates visual heuristics with depth maps—incorporating tissue masking, non-specular filtering, and structural saliency—to guide Gaussian primitive initialization and density control. Furthermore, a temporal texture-aware term is introduced to dynamically weight primitive contributions, enabling robust real-time reconstruction. Experimental evaluations on the EndoNeRF and SCARED benchmarks demonstrate that the proposed approach reduces optical flow error by approximately 27%, successfully achieving high-fidelity rendering while maintaining real-time inference speed.

Abstract

Dynamic endoscopic reconstruction is fundamental to robotic surgery and computer-assisted interventions. While 3D Gaussian Splatting (3DGS) realises real-time rendering, its application to deformable intraoperative environments remains constrained by spurious geometry and varying illuminations. To address these limitations, we introduce EndoPrior-GS, a novel pipeline that explicitly couples frame-extracted vision heuristics and estimated depth maps. EndoPrior-GS derives a joint texture prior from a tool-filtered valid tissue mask, a non-specular photometric filter, and anatomical structural salience, yielding a probability map that guides primitive initialisation and subsequent density control. The prior is further extended to the temporal domain through a texture-aware term that dynamically weighs pairwise primitive contributions during training. We conduct extensive experiments on benchmark datasets EndoNeRF and SCARED, and the obtained results show that our method EndoPrior-GS reduces Flow Error by 27.7% and 25.8% over the representative approaches while preserving competitive rendering quality and real-time rendering speed. Our project website is available at https://jiaqi-huang-77.github.io/EndoPrior-GS/.

Problem

• Dynamic endoscopic reconstruction
• 3D Gaussian Splatting
• Deformable environments
• Spurious geometry
• Varying illumination

Innovation

• 3D Gaussian Splatting
• Dynamic Endoscopic Reconstruction
• Joint Texture Prior
• Depth Estimation
• Robotic Surgery