Surfel-based RGB-D Reconstruction and SLAM with Global and Local Consistency

Abstract

Achieving high surface reconstruction accuracy in dense mapping has been a desirable target for both robotics and vision communities. In the robotics literature, simultaneous localization and mapping (SLAM) systems use depth-enabled cameras to reconstruct a dense map of the environment. They leverage the depth input to provide accurate local pose estimation and a locally consistent model. However, drift in the pose tracking over time leads to misalignments and artifacts. On the other hand, offline computer vision methods, such as the pipeline that combines structure-from-motion (SfM) and multi-view stereo (MVS), estimate the camera poses by performing batch optimization. These methods achieve global consistency, but suffer from heavy computation loads. We propose two novel approaches that integrate both methods to achieve locally and globally consistent reconstruction. The first method estimates poses of the keyframes in the offline SfM pipeline to provide strong global constraints at relatively low cost. Afterwards, we compute odometry between frames driven by off-the-shelf SLAM systems with high local accuracy. We fuse the two pose estimations using factor graph optimization to generate accurate camera poses for dense reconstruction. The second method applies bundle adjustment to improve the estimation of both camera tracking and landmarks, while simultaneously optimizing the dense model upon loop closure using a deformation graph. Through efficient implementation on GPU, the system is able to achieve online performance and accurate dense reconstruction. Experiments on real-world and synthetic datasets demonstrate that our approaches produce more accurate models comparing to existing dense SLAM systems, while achieving significant speedup with respect to state-of-the-art SfM-MVS pipelines.