Advanced search
1 file | 1.02 MB

Cell-based approach for 3D reconstruction from incomplete silhouettes

Author
Organization
Abstract
Shape-from-silhouettes is a widely adopted approach to compute accurate 3D reconstructions of people or objects in a multi-camera environment. However, such algorithms are traditionally very sensitive to errors in the silhouettes due to imperfect foreground-background estimation or occluding objects appearing in front of the object of interest. We propose a novel algorithm that is able to still provide high quality reconstruction from incomplete silhouettes. At the core of the method is the partitioning of reconstruction space in cells, i.e. regions with uniform camera and silhouette coverage properties. A set of rules is proposed to iteratively add cells to the reconstruction based on their potential to explain discrepancies between silhouettes in different cameras. Experimental analysis shows significantly improved F1-scores over standard leave-M-out reconstruction techniques.
Keywords
shape-from-silhouettes, 3D reconstruction, occlusion, multi-camera

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 1.02 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Slembrouck, Maarten, Peter Veelaert, David Van Hamme, Dimitri Van Cauwelaert, and Wilfried Philips. 2017. “Cell-based Approach for 3D Reconstruction from Incomplete Silhouettes.” In Lecture Notes in Computer Science. Springer.
APA
Slembrouck, M., Veelaert, P., Van Hamme, D., Van Cauwelaert, D., & Philips, W. (2017). Cell-based approach for 3D reconstruction from incomplete silhouettes. Lecture Notes in Computer Science. Presented at the Advanced Concepts for Intelligent Vision Systems (ACIVS), Springer.
Vancouver
1.
Slembrouck M, Veelaert P, Van Hamme D, Van Cauwelaert D, Philips W. Cell-based approach for 3D reconstruction from incomplete silhouettes. Lecture Notes in Computer Science. Springer; 2017.
MLA
Slembrouck, Maarten, Peter Veelaert, David Van Hamme, et al. “Cell-based Approach for 3D Reconstruction from Incomplete Silhouettes.” Lecture Notes in Computer Science. Springer, 2017. Print.
@inproceedings{8534000,
  abstract     = {Shape-from-silhouettes is a widely adopted approach to compute accurate 3D reconstructions of people or objects in a multi-camera environment. However, such algorithms are traditionally very sensitive to errors in the silhouettes due to imperfect foreground-background estimation or occluding objects appearing in front of the object of interest. We propose a novel algorithm that is able to still provide high quality reconstruction from incomplete silhouettes. At the core of the method is the partitioning of reconstruction space in cells, i.e. regions with uniform camera and silhouette coverage properties. A set of rules is proposed to iteratively add cells to the reconstruction based on their potential to explain discrepancies between silhouettes in different cameras. Experimental analysis shows significantly improved F1-scores over standard leave-M-out reconstruction techniques.},
  author       = {Slembrouck, Maarten and Veelaert, Peter and Van Hamme, David and Van Cauwelaert, Dimitri and Philips, Wilfried},
  booktitle    = {Lecture Notes in Computer Science},
  keyword      = {shape-from-silhouettes,3D reconstruction,occlusion,multi-camera},
  language     = {eng},
  location     = {Antwerp},
  publisher    = {Springer},
  title        = {Cell-based approach for 3D reconstruction from incomplete silhouettes},
  year         = {2017},
}