User: Guest  Login
Title:

Semantic Monocular SLAM for Highly Dynamic Environments

Document type:
Konferenzbeitrag
Author(s):
Brasch, N.; Bozic, A.; Lallemand, J.; Tombari, F.
Abstract:
Recent advances in monocular SLAM have enabled real-time capable systems which run robustly under the assumption of static environment, but fail in presence of dynamic scene changes and motion, since they lack an explicit dynamic outlier handling. We propose a semantic monocular SLAM framework designed to deal with highly dynamic environments, combining feature-based and direct approaches to achieve robustness under challenging conditions. The proposed approach exploits semantic information ext...     »
Keywords:
cameras; feature extraction; image motion analysis; image sequences; mobile robots; object detection; object tracking; pose estimation; probability; robot vision; SLAM (robots); static environment; semantic monocular SLAM framework; semantic information; explicit probabilistic model; dynamic environments; Virtual KITTI; Synthia datasets; pose estimation; Semantics; Simultaneous localization and mapping; Feature extraction; Dynamics; Cameras; Pose estimation; Probabilistic logic
Book / Congress title:
2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
Year:
2018
Pages:
393--400
 BibTeX