SAGS: Self-Adaptive Alias-Free Gaussian Splatting for Dynamic Surgical Endoscopic Reconstruction
🤖 AI Summary
Dynamic endoscopic videos suffer from aliasing and artifacts due to tissue motion, yet existing 3D Gaussian Splatting (3DGS) methods prioritize rendering speed at the expense of reconstruction fidelity. To address this, we propose an adaptive anti-aliased Gaussian splatting framework. Our method introduces an attention-driven dynamic weighting 4D deformation decoder—the first to incorporate adaptive filtering into 4D Gaussian representations—and jointly integrates 3D spatial smoothing with 2D Mip-mapping to model soft-tissue non-rigid motion without aliasing. This design effectively suppresses deformation artifacts while enhancing temporal consistency and fine-detail recovery. Evaluated on EndoNeRF and SCARED benchmarks, our approach achieves state-of-the-art performance across PSNR, SSIM, and LPIPS metrics. Qualitative results demonstrate superior visual realism and temporal stability, confirming both technical advancement and clinical applicability.
Application Category
📝 Abstract
Surgical reconstruction of dynamic tissues from endoscopic videos is a crucial technology in robot-assisted surgery. The development of Neural Radiance Fields (NeRFs) has greatly advanced deformable tissue reconstruction, achieving high-quality results from video and image sequences. However, reconstructing deformable endoscopic scenes remains challenging due to aliasing and artifacts caused by tissue movement, which can significantly degrade visualization quality. The introduction of 3D Gaussian Splatting (3DGS) has improved reconstruction efficiency by enabling a faster rendering pipeline. Nevertheless, existing 3DGS methods often prioritize rendering speed while neglecting these critical issues. To address these challenges, we propose SAGS, a self-adaptive alias-free Gaussian splatting framework. We introduce an attention-driven, dynamically weighted 4D deformation decoder, leveraging 3D smoothing filters and 2D Mip filters to mitigate artifacts in deformable tissue reconstruction and better capture the fine details of tissue movement. Experimental results on two public benchmarks, EndoNeRF and SCARED, demonstrate that our method achieves superior performance in all metrics of PSNR, SSIM, and LPIPS compared to the state of the art while also delivering better visualization quality.
Problem
Research questions and friction points this paper is trying to address.
Addresses aliasing and artifacts in endoscopic tissue reconstruction
Improves dynamic surgical scene reconstruction quality
Enhances fine detail capture in deformable tissue movement
Innovation
Methods, ideas, or system contributions that make the work stand out.
Self-adaptive alias-free Gaussian splatting framework
Attention-driven dynamically weighted 4D deformation decoder
3D smoothing filters and 2D Mip filters integration
🔎 Similar Papers
No similar papers found.
