Coloured signed distance fields for full 3D object reconstruction

Kehl W, Navab N, Ilic S (2014)

Publication Type: Conference contribution

Publication year: 2014

Publisher: British Machine Vision Association, BMVA

Conference Proceedings Title: BMVC 2014 - Proceedings of the British Machine Vision Conference 2014

Event location: Nottingham, GBR

Abstract

We propose a novel 3D object reconstruction framework that is able to fully capture the accurate coloured geometry of an object using an RGB-D sensor. Building on visual odometry for trajectory estimation, we perform pose graph optimisation on collected keyframes and reconstruct the scan variationally via coloured signed distance fields. To capture the full geometry, we conduct multiple scans while changing the object's pose. After collecting all coloured fields, we perform an automated dense registration over all collected scans to create one coherent model. We show on eight reconstructed real-life objects that the proposed pipeline outperforms the state-of-the-art in visual quality as well as geometrical fidelity.

Involved external institutions

Technische Universität München (TUM)

Germany (DE)

How to cite

APA:

Kehl, W., Navab, N., & Ilic, S. (2014). Coloured signed distance fields for full 3D object reconstruction. In Michel Valstar, Andrew French, Tony Pridmore (Eds.), BMVC 2014 - Proceedings of the British Machine Vision Conference 2014. Nottingham, GBR: British Machine Vision Association, BMVA.

MLA:

Kehl, Wadim, Nassir Navab, and Slobodan Ilic. "Coloured signed distance fields for full 3D object reconstruction." Proceedings of the 25th British Machine Vision Conference, BMVC 2014, Nottingham, GBR Ed. Michel Valstar, Andrew French, Tony Pridmore, British Machine Vision Association, BMVA, 2014.

BibTeX: Download