Mapping and localization from planar markers

作者：

Highlights：

• Cost-effective approach for large-scale camera localization using planar squared fiducial markers

• Marker mapping is tackled as a variant of Sparse Bundle Adjustment problem jointly optimizing the four marker corners.

• A graph-based method to obtain the initial map of markers dealing with the ambiguity problem is proposed.

• A localization method considering all visible markers, which is able to cope with the ambiguity problem is proposed.

摘要

•Cost-effective approach for large-scale camera localization using planar squared fiducial markers•Marker mapping is tackled as a variant of Sparse Bundle Adjustment problem jointly optimizing the four marker corners.•A graph-based method to obtain the initial map of markers dealing with the ambiguity problem is proposed.•A localization method considering all visible markers, which is able to cope with the ambiguity problem is proposed.

论文关键词：Fiducial markers,Marker mapping,SLAM,SfM

论文评审过程：Received 30 January 2017, Revised 13 July 2017, Accepted 6 August 2017, Available online 18 August 2017, Version of Record 18 September 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2017.08.010