EndoMD-SLAM: Endoscopic Gaussian Splatting SLAM under Optical Degradation with Memory and Static-Transient Decomposition

2026-08-09

AI Summary

This work addresses the challenges of endoscopic visual SLAM under optical degradations such as moving debris and fluid flow, which violate photometric consistency assumptions and lead to map corruption and tracking drift. The authors propose a memory-augmented SLAM framework that employs a memory-driven tracking gating mechanism to suspend map updates during unreliable observations and leverages historical keyframes for robust relocalization. Additionally, a self-supervised static–transient scene decomposition module isolates dynamic contaminants into a transient field, thereby preserving the integrity of anatomical structure maps. This approach uniquely integrates static–transient decomposition with memory-enhanced relocalization, achieving a 91% reduction in absolute trajectory error and a 9.9 dB improvement in rendering PSNR on a colonoscopy degradation dataset, significantly outperforming existing baselines.

Abstract

Dense 3D reconstruction is critical for clinical endoscopic navigation and documentation. While Gaussian Splatting SLAM systems show promise in this domain, they fundamentally rely on strict multi-view photometric consistency. In routine procedures, this assumption is severely violated by intermittent optical degradations like moving debris and water flushing. Standard systems erroneously fuse these cameraattached artifacts into the persistent 3D geometry, causing severe tracking drift and irreversible map corruption. To address this limitation, we propose EndoMD-SLAM, a framework designed to maintain stability under optical degradation through specialized tracking and mapping mechanisms. On the tracking side, a memory-driven gating mechanism detects unreliable observations to suspend map updates and utilizes historical keyframes for drift-aware relocalization. On the mapping side, a self-supervised static-transient decomposition isolates visual contaminants into a dedicated transient field. This explicit separation prevents artifacts from structurally entangling with the persistent anatomical map. We curate a degradationfocused benchmark from colonoscopy videos to systematically evaluate these failure modes. Extensive experiments show that while standard baselines fail under severe optical degradation, EndoMD-SLAM preserves geometric integrity, reducing absolute trajectory error by 91% and improving rendering fidelity by 9.9 dB PSNR.

Problem

• optical degradation
• endoscopic SLAM
• photometric inconsistency
• map corruption
• visual artifacts

Innovation

• Gaussian Splatting SLAM
• optical degradation
• static-transient decomposition
• memory-driven tracking
• endoscopic reconstruction