DualRefine: Self-Supervised Depth and Pose Estimation Through Iterative Epipolar Sampling and Refinement Toward Equilibrium

arXiv:2304.03560v2 Announce Type: replace
Abstract: Self-supervised multi-frame depth estimation achieves high accuracy by computing matching costs of pixel correspondences between adjacent frames, injecting geometric information into the network. These pixel-correspondence candidates are computed based on the relative pose estimates between the frames. Accurate pose predictions are essential for precise matching cost computation as they influence the epipolar geometry. Furthermore, improved depth estimates can, in turn, be used to align pose estimates.
Inspired by traditional structure-from-motion (SfM) principles, we propose the …

abstract accuracy arxiv computing costs cs.cv equilibrium information iterative network pixel predictions sampling through type