VOLDOR-SLAM: For the Times When Feature-Based or Direct Methods Are Not Good Enough

TL;DR

VOLDOR-SLAM introduces a dense-indirect SLAM system that utilizes dense optical flows and geometric priors to improve robustness and accuracy in monocular, stereo, and RGB-D scenarios, enabling real-time dense mapping.

Contribution

The paper extends the VOLDOR probabilistic visual odometry model by integrating geometric priors and a novel pose graph management scheme for enhanced dense SLAM performance.

Findings

Achieves accurate and robust camera pose estimation using dense optical flow.

Supports monocular, stereo, and RGB-D input seamlessly.

Operates online at 15 FPS on a GTX1080Ti GPU.

Abstract

We present a dense-indirect SLAM system using external dense optical flows as input. We extend the recent probabilistic visual odometry model VOLDOR [Min et al. CVPR'20], by incorporating the use of geometric priors to 1) robustly bootstrap estimation from monocular capture, while 2) seamlessly supporting stereo and/or RGB-D input imagery. Our customized back-end tightly couples our intermediate geometric estimates with an adaptive priority scheme managing the connectivity of an incremental pose graph. We leverage recent advances in dense optical flow methods to achieve accurate and robust camera pose estimates, while constructing fine-grain globally-consistent dense environmental maps. Our open source implementation [https://github.com/htkseason/VOLDOR] operates online at around 15 FPS on a single GTX1080Ti GPU.