BASED: Bundle-Adjusting Surgical Endoscopic Dynamic Video Reconstruction Using Neural Radiance Fields
Year: 2025
Type: conference-paper
Abstract: Reconstruction of deformable scenes from endoscopic videos is important for many applications such as intraop-erative navigation, surgical visual perception, and robotic surgery. It is a foundational requirement for realizing autonomous robotic interventions for minimally invasive surgery. However, previous approaches in this domain have been limited by their modular nature and are confined to specific camera and scene settings. Our work adopts the Neural Radiance Fields (NeRF) approach to learning 3D implicit representations of scenes that are both dynamic and deformable over time, and furthermore with unknown cam-era poses. This work removes the constraints of known cam-era poses and overcomes the drawbacks of the state-of-the-art unstructured dynamic scene reconstruction technique, which relies on the static part of the scene for accurate re-construction. Through several experimental datasets, we demonstrate the versatility of our proposed model to adapt to diverse camera and scene settings, and show its promise for both current and future robotic surgical systems.
Authors: Shreya Saha, Zekai Liang, Shan Lin, Jingpei Lu, Michael Yip, Sainan Liu
Institutions: University of California San Diego, Intel (United States)
Language: English
Subfield: Computational Mechanics
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4409263065