NeRFscopy: Neural Radiance Fields for in-vivo Time-Varying Tissues from Endoscopy
This addresses the problem of enhancing visualization and diagnostic accuracy in medical endoscopy, representing a domain-specific incremental advance.
The paper tackles 3D reconstruction of deformable tissues from endoscopic videos by introducing NeRFscopy, a self-supervised pipeline that achieves accurate novel view synthesis and outperforms competing methods on challenging endoscopy scenes.
Endoscopy is essential in medical imaging, used for diagnosis, prognosis and treatment. Developing a robust dynamic 3D reconstruction pipeline for endoscopic videos could enhance visualization, improve diagnostic accuracy, aid in treatment planning, and guide surgery procedures. However, challenges arise due to the deformable nature of the tissues, the use of monocular cameras, illumination changes, occlusions and unknown camera trajectories. Inspired by neural rendering, we introduce NeRFscopy, a self-supervised pipeline for novel view synthesis and 3D reconstruction of deformable endoscopic tissues from a monocular video. NeRFscopy includes a deformable model with a canonical radiance field and a time-dependent deformation field parameterized by SE(3) transformations. In addition, the color images are efficiently exploited by introducing sophisticated terms to learn a 3D implicit model without assuming any template or pre-trained model, solely from data. NeRFscopy achieves accurate results in terms of novel view synthesis, outperforming competing methods across various challenging endoscopy scenes.