ESTHER: Joint Camera Self-Calibration and Automatic Radial Distortion Correction from Tracking of Walking Humans
IEEE Access
Summary
ESTHER jointly estimates camera parameters and radial distortion from tracked walking people. Tracking-assisted segmentation supplies head and foot observations; mean-shift clustering and Laplace regression estimate vanishing points. An evolutionary optimization method then refines camera and distortion parameters together. Evaluations on public benchmarks and captured video examine calibration robustness and the resulting conversion of 2D object tracks into 3D scene coordinates.
Citation
@article{Tang19ESTHER,
author = {Zheng Tang and Yen-Shuo Lin and Kuan-Hui Lee and Jenq-Neng Hwang},
title = { {ESTHER}: {J}oint camera self-calibration and automatic radial distortion correction from tracking of walking humans},
journal = {IEEE Access},
volume = {7},
pages = {10754--10766},
year = {2019}
}
