Search papers, labs, and topics across Lattice.
This study introduces Endo-NeRF++, a neural rendering framework designed to reconstruct dynamic surgical scenes by addressing challenges such as tissue deformation and occlusions. By integrating multi-resolution hash-grid encoding, temporal feature merging, and uncertainty-informed adaptive sampling, the framework significantly enhances both reconstruction accuracy and temporal coherence. Experimental results show that Endo-NeRF++ improves PSNR by up to 1.22 dB, SSIM by up to 5.3%, and reduces LPIPS by up to 55.1% compared to its predecessor, EndoNeRF.
Uncertainty-aware adaptive sampling in Endo-NeRF++ boosts rendering quality in dynamic surgical scenes, achieving significant improvements over previous methods.
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.
