FlowMap: High-Quality Camera Poses, Intrinsics, and Depth via Gradient Descent

arXiv:2404.15259v1 Announce Type: new
Abstract: This paper introduces FlowMap, an end-to-end differentiable method that solves for precise camera poses, camera intrinsics, and per-frame dense depth of a video sequence. Our method performs per-video gradient-descent minimization of a simple least-squares objective that compares the optical flow induced by depth, intrinsics, and poses against correspondences obtained via off-the-shelf optical flow and point tracking. Alongside the use of point tracks to encourage long-term geometric consistency, we introduce differentiable re-parameterizations of depth, intrinsics, and …

abstract arxiv cs.cv differentiable flow gradient gradient-descent least optical optical flow paper per quality simple squares type via video