Endo-NeRF++: Uncertainty-Aware Neural Rendering with Multi-Resolution Hash Encoding for Dynamic Surgical Scene Reconstruction

2026-07-30

AI Summary

Dynamic surgical scene reconstruction often suffers from geometric inconsistencies and degraded rendering quality due to tissue deformation, occlusions, specular reflections, and limited viewpoints. To address these challenges, this work proposes Endo-NeRF++, which innovatively integrates multi-resolution hash encoding, temporal feature aggregation, and an uncertainty-aware adaptive ray sampling mechanism. This approach significantly enhances temporal consistency and reconstruction accuracy of neural radiance fields in endoscopic dynamic scenes. Experiments on robotic surgery videos demonstrate that Endo-NeRF++ outperforms the EndoNeRF baseline, achieving a maximum PSNR gain of 1.22 dB (+4.3%), a 5.3% improvement in SSIM, and a 55.1% reduction in LPIPS.

Abstract

Reconstructing dynamic surgical scenes is crucial for robot-assisted minimally invasive surgery; however, it continues to be difficult because of tissue deformation, occlusions, specular reflections, and restricted viewpoints. In this study, we introduce Endo-NeRF++, a neural rendering framework that accounts for uncertainty in the reconstruction of dynamic surgical scenes. Expanding on EndoNeRF, the suggested approach incorporates multi-resolution hash-grid encoding, temporal feature merging, and uncertainty-informed adaptive sampling to enhance reconstruction accuracy and temporal coherence in deformable endoscopic scenes.The multi-resolution hash-grid representation within the framework effectively captures both coarse and fine anatomical details, while temporal feature blending ensures stable reconstruction during tissue deformation and surgical tool occlusions. Additionally, uncertainty-driven adaptive sampling assigns more samples to uncertain areas to enhance rendering quality and geometric coherence. Experiments on robotic surgical video sequences demonstrate that the proposed uncertainty-guided adaptive sampling improves PSNR by up to 1.22\,dB (4.3\%), increases SSIM by up to 5.3\%, and reduces LPIPS by up to 55.1\% compared with the EndoNeRF baseline.

Problem

• dynamic surgical scene reconstruction
• tissue deformation
• occlusions
• specular reflections
• restricted viewpoints

Innovation

• uncertainty-aware rendering
• multi-resolution hash encoding
• temporal feature blending
• adaptive sampling
• dynamic surgical scene reconstruction