On the Uncertain Single-View Depths in Colonoscopies
J. Rodríguez-Puigvert*, D. Recasens , R. Martinez-Cantin, J. Civera
MICCAI 2022
Abstract
Estimating depth information from endoscopic images is a prerequisite for a wide set of AI-assisted technologies, such as accurate localization and measurement of tumors, or identification of non-inspected areas. As the domain specificity of colonoscopies -- deformable low-texture environments with fluids, poor lighting conditions and abrupt sensor motions -- pose challenges to multi-view 3D reconstructions, single-view depth learning stands out as a promising line of research. Depth learning can be extended in a Bayesian setting, which enables continual learning, improves decision making and can be used to compute confidence intervals or quantify uncertainty for in-body measurements. In this paper, we explore for the first time Bayesian deep networks for single-view depth estimation in colonoscopies. Our specific contribution is two-fold: 1) an exhaustive analysis of scalable Bayesian networks for depth learning in different datasets, highlighting challenges and conclusions regarding synthetic-to-real domain changes and supervised vs. self-supervised methods; and 2) a novel teacher-student approach to deep depth learning that takes into account the teacher uncertainty.
Architecture:
Citation:
@inproceedings{rodriguez2022uncertain,
title={On the uncertain single-view depths in colonoscopies},
author={Rodriguez-Puigvert, Javier and Recasens, David and Civera, Javier and Martinez-Cantin, Ruben},
booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
pages={130--140},
year={2022},
organization={Springer}
}