A Compact Parametric Solution to Depth Sensor Calibration

Andrew Spek and Tom Drummond

Abstract

We present a method for calibration of low-cost depth sensors such as the Microsoft Kinect. We show this method is effective at correcting the structured sensor error using a compact parametric solution, that uses only a small fraction of the number of parameters used in many existing approaches. Additionally we provide this calibration as an open-source implementation, with limited required external dependencies. We demonstrate our approach can optimise directly for a geometric depth and radial distortion calibration function in minutes on modern hardware. We demonstrate our approach can optimise directly for a geometric depth and radial distortion calibration function in minutes on modern hardware.

Session

Posters

Files

PDF iconPaper (PDF)

DOI

10.5244/C.31.179
https://dx.doi.org/10.5244/C.31.179

Citation

Andrew Spek and Tom Drummond. A Compact Parametric Solution to Depth Sensor Calibration. In T.K. Kim, S. Zafeiriou, G. Brostow and K. Mikolajczyk, editors, Proceedings of the British Machine Vision Conference (BMVC), pages 179.1-179.12. BMVA Press, September 2017.

Bibtex

            @inproceedings{BMVC2017_179,
                title={A Compact Parametric Solution to Depth Sensor Calibration},
                author={Andrew Spek and Tom Drummond},
                year={2017},
                month={September},
                pages={179.1-179.12},
                articleno={179},
                numpages={12},
                booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
                publisher={BMVA Press},
                editor={Tae-Kyun Kim, Stefanos Zafeiriou, Gabriel Brostow and Krystian Mikolajczyk},
                doi={10.5244/C.31.179},
                isbn={1-901725-60-X},
                url={https://dx.doi.org/10.5244/C.31.179}
            }