Structure from Motion - ToF-aided 3D Reconstruction of Isometric Surfaces
S. Jafar Hosseini, Helder Araújo
2014
Abstract
This paper deals with structure-from-motion (SfM) for non-rigid surfaces that undergo isometric motion. Our SfM framework aims at the joint estimation of the 3D surface and the camera motion by combining a ToF range sensor and a monocular RGB camera through a template-based approach. Our goal is to use the 2D low-resolution depth estimates provided by the TOF camera, in order to facilitate the estimation of non-rigid structure using the high-resolution images obtained by means of a RGB camera. In this paper, we model isometric surfaces with a triangular mesh. The ToF sensor is used to obtain the depth of a sparse set of 3D feature points, from which the depth of the mesh vertices can be recovered using a multivariate linear system. Subsequently, we form a non-linear constraint based on the projected length of each edge. A second non-linear constraint is then used for minimizing re-projection errors. These constraints are finally incorporated into an optimization scheme to solve for structure and motion. Experimental results show that the proposed approach has good performance even if only a low-resolution depth image is used.
References
- Aans, H. and Kahl, F. (2002). Estimation of deformable structure and motion. Workshop on Vision and Modelling of Dynamic Scenes, ECCV, Denmark.
- Akhter, I., Sheikh, Y., and Khan, S. (2009). In defense of orthonormality constraints for nonrigid structure from motion. pages 1534-1541. CVPR.
- Dai, Y., Li, H., and He, M. (2012). A simple prior-free method for non-rigid structure-from-motion factorization. pages 2018-2025. CVPR.
- Del-Bue, A., Llad, X., and Agapito, L. (2006). Non-rigid metric shape and motion recovery from uncalibrated images using priors. IEEE Conference on Computer Vision and Pattern Recognition, New York.
- Diebel, J. and Thrun, S. (2005). An application of markov random fields to range sensing. Proc. NIPS.
- Gay-Bellile, V., Perriollat, M., Bartoli, A., and Sayd, P. (2006). Image registration by combining thin-plate splines with a 3d morphable model. International Conference on Image Processing.
- Gumerov, N., Zandifar, A., Duraiswami, R., and Davis, L. (2004). Structure of applicable surfaces from single views. European Conference on Computer Vision.
- Kim, H., Tai, Y.-W., and Brown, M. (2011). High quality depth map upsampling for 3d-tof cameras. pages 1623 - 1630. Inso Kweon Computer Vision (ICCV), IEEE International Conference, Barcelona.
- Kim, Y., Theobalt, C., Diebel, J., Kosecka, J., Miscusik, B., and Thrun, S. (2009). Multi-view image and tof sensor fusion for dense 3d reconstruction. pages 1542-1549. Computer Vision Workshops (ICCV Workshops).
- Llado, X., Bue, A., and Agapito, L. (2005). Non-rigid 3d factorization for projective reconstruction. BMVC.
- Metaxas, D. and Terzopoulos, D. (1993). Constrained deformable superquadrics and nonrigid motion tracking. PAMI 15, pages 580-591.
- Paladini, M., Bue, A., Stosic, M., Dodig, M., Xavier, J., and Agapito, L. (2009). Factorization for non-rigid and articulated structure using metric projections. page 28982905. Proc. IEEE Conf. on Computer Vision and Pattern Recognition.
- Perriollat, M., Hartley, R., and Bartoli (2010). Monocular template-based reconstruction of inextensible surfaces. International Journal of Computer Vision.
- Prasad, M., Zisserman, A., and Fitzgibbon, A. (2006). Single view reconstruction of curved surfaces. pages 1345-1354. IEEE Conference on Computer Vision and Pattern Recognition.
- Salzmann, M. and Fua, P. (2007). Reconstructing sharply folding surfaces: A convex formulation. IEEE Conference on Computer Vision and Pattern Recognition.
- Salzmann, M., Hartley, R., and Fua, P. (2007). Convex optimization for deformable surface 3-d tracking. IEEE International Conference on Computer Vision.
- Salzmann, M., Moreno-Noguer, F., Lepetit, V., and Fua, P. (2008). Closed-form solution to non-rigid 3d surface registration. pages 581-594. European Conference on Computer Vision.
- Shen, S., Shi, W., and Liu, Y. (2010). Monocular 3-d tracking of inextensible deformable surfaces under l2- norm. IEEE Transactions on Image Processing 19, pages 512-521.
- Srivastava, S., Saxena, A., Theobalt, C., and Thrun, S. (2009). Rapid interactive 3d reconstruction from a single image. In VMV, pages 19-28.
- Torresani, L., Hertzmann, A., and Bregler, C. (2003). Learning non-rigid 3d shape from 2d motion. NIPS, pages 580-591.
- White, R. and Forsyth, D. (2006). Combining cues: Shape from shading and texture. CVPR.
- Xiao, J., x. Chai, J., and Kanade, T. (2004). A closedform solution to non-rigid shape and motion recovery. pages 573-587. ECCV.
- Yang, R., Davis, J., and Nister, D. (2007). Spatial-depth super resolution for range images. pages 1-8. Computer Vision and Pattern Recognition, CVPR 7807. IEEE Conference, Minneapolis, MN.
- Zhou, H., Li, X., and Sadka, A. (2012). Nonrigid structurefrom-motion from 2-d images using markov chain monte carlo. 14(1):168-177.
Paper Citation
in Harvard Style
Hosseini S. and Araújo H. (2014). Structure from Motion - ToF-aided 3D Reconstruction of Isometric Surfaces . In Proceedings of the 3rd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-018-5, pages 544-552. DOI: 10.5220/0004787505440552
in Bibtex Style
@conference{icpram14,
author={S. Jafar Hosseini and Helder Araújo},
title={Structure from Motion - ToF-aided 3D Reconstruction of Isometric Surfaces},
booktitle={Proceedings of the 3rd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2014},
pages={544-552},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004787505440552},
isbn={978-989-758-018-5},
}
in EndNote Style
TY - CONF
JO - Proceedings of the 3rd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Structure from Motion - ToF-aided 3D Reconstruction of Isometric Surfaces
SN - 978-989-758-018-5
AU - Hosseini S.
AU - Araújo H.
PY - 2014
SP - 544
EP - 552
DO - 10.5220/0004787505440552