Useful structure constraints in indoor SLAM systems


Details
Indoor environments offer, in addition to points, an abundance of geometrical features such as lines and planes, which we exploit to design both the tracking and mapping components of our SLAM system. For the tracking part, we explore geometric relationships between these features to achieve accurate camera pose estimation performances. For the mapping part, different levels of maps from sparse to dense are reconstructed at a low computational cost.
Lecture slides: https://docs.google.com/presentation/d/1X2M9iZz9_LXnOtfX5HXTrDinNXbK_ex04hcJn9hpPYo/edit?usp=sharing
The talk is based on the speaker's papers:
Structure-SLAM: Low-Drift Monocular SLAM in Indoor Environments (RAL/IROS2020)
https://arxiv.org/pdf/2008.01963.pdf
RGB-D SLAM with Structural Regularities (ICRA2021), https://arxiv.org/pdf/2010.07997.pdf
code: https://github.com/yanyan-li/PlanarSLAM
Presenter BIO:
Mr. Yanyan Li is currently a Ph.D. candidate in Technical University of Munich (TUM), he received his B.Eng. and M.Eng Degree from Anhui Polytechnic University and Peking University, respectively. Yanyan's research interests include multi-view geometry and neural networks, where more information about him can be found at http://campar.in.tum.de/Main/YanyanLi

Useful structure constraints in indoor SLAM systems