Online 3D reconstruction and dense tracking in endoscopic videos
This work addresses the problem of real-time surgical scene understanding for medical professionals, though it appears incremental as it builds on existing reconstruction techniques.
The researchers tackled 3D scene reconstruction from stereo endoscopic video for surgical interventions by developing an online framework using Gaussian splatting and control points for tissue deformation, achieving state-of-the-art tracking performance and comparable results to offline methods on the StereoMIS dataset.
3D scene reconstruction from stereo endoscopic video data is crucial for advancing surgical interventions. In this work, we present an online framework for online, dense 3D scene reconstruction and tracking, aimed at enhancing surgical scene understanding and assisting interventions. Our method dynamically extends a canonical scene representation using Gaussian splatting, while modeling tissue deformations through a sparse set of control points. We introduce an efficient online fitting algorithm that optimizes the scene parameters, enabling consistent tracking and accurate reconstruction. Through experiments on the StereoMIS dataset, we demonstrate the effectiveness of our approach, outperforming state-of-the-art tracking methods and achieving comparable performance to offline reconstruction techniques. Our work enables various downstream applications thus contributing to advancing the capabilities of surgical assistance systems.