Enabling Geometry Aware Learning Through Differentiable Epipolar View Translation

Abstract

Epipolar geometry is exploited in several applications in the field of Cone-Beam Computed Tomography (CBCT) imaging. By leveraging consistency conditions between multiple views of the same scene, motion artifacts can be minimized, the effects of beam hardening can be reduced, and segmentation masks can be refined. In this work, we explore the idea of enabling deep learning models to access the known geometrical relations between views. This implicit 3D information can potentially enhance various projection domain algorithms such as segmentation, detection, or inpainting. We introduce a differentiable feature translation operator, which uses available projection matrices to calculate and integrate over the epipolar line in a second view. As an example application, we evaluate the effects of the operator on the task of projection domain metal segmentation. By re-sampling a stack of projections into orthogonal view pairs, we segment each projection image jointly with a second view acquired roughly 90$^\circ$ apart. The comparison with an equivalent single-view segmentation model reveals an improved segmentation performance of 0.95 over 0.91 measured by the dice coefficient. By providing an implementation of this operator as an open-access differentiable layer, we seek to enable future research.