Details
| Original language | English |
|---|---|
| Title of host publication | Computer Vision |
| Subtitle of host publication | ACCV 2016 Workshop |
| Publisher | Springer Verlag |
| Pages | 61-75 |
| Number of pages | 15 |
| ISBN (print) | 9783319545257 |
| Publication status | Published - 16 Mar 2017 |
| Event | 13th Asian Conference on Computer Vision, ACCV 2016 - Taipei, Taiwan Duration: 20 Nov 2016 → 24 Nov 2016 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 10118 LNCS |
| ISSN (Print) | 0302-9743 |
| ISSN (electronic) | 1611-3349 |
Abstract
This paper discusses the efficient optimization of iterative closest points (ICP) algorithms. While many algorithms formulate the optimization problem in terms of quadratic error functionals, the discontinuities introduced by varying changing correspondences usually motivate the optimization by quasi-Newton or Gauss-Newton methods. These disregard the fact that the Hessian matrix in these cases is constant, and can thus be precomputed analytically and inverted a-priori. We demonstrate on the example of Allen et al.’s seminal paper “The space of human body shapes”, that all relevant quantities for a full Newton method can be derived easily, and lead to an optimization process that reduces computation time by around 98% while achieving results of almost equal quality (about 1% difference). Along the way, the paper proposes minor improvements to the original problem formulation by Allen et al., aimed at making the results more reproducible.
ASJC Scopus subject areas
- Mathematics(all)
- Theoretical Computer Science
- Computer Science(all)
- General Computer Science
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
Computer Vision: ACCV 2016 Workshop. Springer Verlag, 2017. p. 61-75 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10118 LNCS).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Rapid analytic optimization of quadratic ICP algorithms
AU - German, Leonid
AU - Ziehn, Jens
AU - Rosenhahn, Bodo
PY - 2017/3/16
Y1 - 2017/3/16
N2 - This paper discusses the efficient optimization of iterative closest points (ICP) algorithms. While many algorithms formulate the optimization problem in terms of quadratic error functionals, the discontinuities introduced by varying changing correspondences usually motivate the optimization by quasi-Newton or Gauss-Newton methods. These disregard the fact that the Hessian matrix in these cases is constant, and can thus be precomputed analytically and inverted a-priori. We demonstrate on the example of Allen et al.’s seminal paper “The space of human body shapes”, that all relevant quantities for a full Newton method can be derived easily, and lead to an optimization process that reduces computation time by around 98% while achieving results of almost equal quality (about 1% difference). Along the way, the paper proposes minor improvements to the original problem formulation by Allen et al., aimed at making the results more reproducible.
AB - This paper discusses the efficient optimization of iterative closest points (ICP) algorithms. While many algorithms formulate the optimization problem in terms of quadratic error functionals, the discontinuities introduced by varying changing correspondences usually motivate the optimization by quasi-Newton or Gauss-Newton methods. These disregard the fact that the Hessian matrix in these cases is constant, and can thus be precomputed analytically and inverted a-priori. We demonstrate on the example of Allen et al.’s seminal paper “The space of human body shapes”, that all relevant quantities for a full Newton method can be derived easily, and lead to an optimization process that reduces computation time by around 98% while achieving results of almost equal quality (about 1% difference). Along the way, the paper proposes minor improvements to the original problem formulation by Allen et al., aimed at making the results more reproducible.
UR - http://www.scopus.com/inward/record.url?scp=85016089048&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-54526-4_5
DO - 10.1007/978-3-319-54526-4_5
M3 - Conference contribution
AN - SCOPUS:85016089048
SN - 9783319545257
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 61
EP - 75
BT - Computer Vision
PB - Springer Verlag
T2 - 13th Asian Conference on Computer Vision, ACCV 2016
Y2 - 20 November 2016 through 24 November 2016
ER -