A GPU-based elastic shape registration approach in implicit spaces

In this paper, we present a GPU-based implementation of an elastic shape registration approach in implicit spaces. Shapes are represented using signed distance functions, while deformations are modeled by cubic B-splines. In a variational framework, an incremental free form deformation strategy is a...

Full description

Saved in:
Bibliographic Details
Published in:Journal of real-time image processing Vol. 16; no. 6; pp. 2059 - 2071
Main Authors: Yousef, Ahmed Hassan, Abd El Munim, Hossam E.
Format: Journal Article
Language:English
Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.12.2019
Springer Nature B.V
Subjects:
Adaptive control
Algorithms
B spline functions
Closed form solutions
Computer Graphics
Computer Science
Deformation
Exact solutions
Free form
Image Processing and Computer Vision
Multimedia Information Systems
Original Research Paper
Pattern Recognition
Registration
Signal,Image and Speech Processing
CUDA
Elastic registration
Free form deformation (FFD)
GPU
Distance transform
Shape modeling and representation
ISSN:1861-8200, 1861-8219
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we present a GPU-based implementation of an elastic shape registration approach in implicit spaces. Shapes are represented using signed distance functions, while deformations are modeled by cubic B-splines. In a variational framework, an incremental free form deformation strategy is adopted to handle smooth deformations through an adaptive size control lattice grid. The grid control points are estimated by a closed-form solution which avoids the gradient descent iterations. However, even this solution is very far from real time. We show in detail that such an algorithm is computationally expensive with a time complexity of O ( N C P x N C P 2 X 2 Y 2 ) where N C P x and NCP are the grid lattice resolution parameters in the shape domain of size X × Y . Moreover, the problem becomes more time-consuming with the increase in the number of control points because this requires the execution of the incremental algorithm several times. The closed-form solution was implemented using eight different GPU techniques. Our experimental results demonstrate speedups of more than 150 × compared to the C implementation on a CPU.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1861-8200
1861-8219
DOI:10.1007/s11554-017-0710-7