Camera Pose Estimation Using a VLC-Modulated Single Rectangular LED for Indoor Positioning

Babar Hussain, Yiru Wang, Runzhou Chen, C. Patrick Yue

Abstract: An indoor positioning algorithm is presented that uses a single rectangular visible light communication (VLC)-modulated LED as a transmitter and the CMOS image sensor of a smartphone camera as a receiver. The real-world location of the LED is transmitted as a VLC signal that is decoded by the smartphone using optical camera communication (OCC). Then, the location of the smartphone is calculated by finding the camera pose using the perspective-n-point (PnP) algorithm on the real-world coordinates and corresponding image coordinates of the LED. The algorithm is experimentally evaluated as a real-time application running on a smartphone. The experimental results show that the algorithm can achieve an average 3-D positioning error of less than 5 cm for an arbitrarily tilted smartphone.

External IDs:doi:10.1109/tim.2022.3212980