Pop-up SLAM: a Semantic Monocular Plane SLAM for Low-texture Environments

Existing simultaneous localization and mapping (SLAM) algorithm is not robust in challenging low-texture environments because of few salient features. The resulting sparse or semi-dense map also conveys little information for motion planning. Though some work utilize plane or scene layout for dense map regularization, they require decent state estimation from other sources. In this paper, we propose a real-time monocular plane SLAM to demonstrate that scene understanding could improve both state estimation and dense mapping especially in low-texture environments. The plane measurements come from the popup 3D plane model from each single image. We also combine planes with point based SLAM to solve the ill-constrained problems. On a public TUM dataset, our algorithm generates dense semantic 3D model with pixel depth error of 6.2 cm while existing SLAM fails. On a 60m long dataset with loops, our method creates a much better 3D model with state estimation error of 0.67%.