Authors:
Jens Schneider
;
Martin Kraus
and
Rüdiger Westermann
Affiliation:
Computer Graphics & Visualization Group, Technische Universität München, Germany
Keyword(s):
Discrete euclidean distance transform, Graphics processing unit, SIMD-parallelism.
Related
Ontology
Subjects/Areas/Topics:
Computational Geometry
;
Computer Vision, Visualization and Computer Graphics
;
Image Filtering
;
Image Formation and Preprocessing
;
Implementation of Image and Video Processing Systems
;
Motion, Tracking and Stereo Vision
;
Real-Time Vision
Abstract:
We present a discrete distance transform in style of the vector propagation algorithm by Danielsson. Like other vector propagation algorithms, the proposed method is close to exact, i.e., the error can be strictly bounded from above and is significantly smaller than one pixel. Our contribution is that the algorithm runs entirely on consumer class graphics hardware, thereby achieving a throughput of up to 96 Mpixels/s. This allows the proposed method to be used in a wide range of applications that rely both on high speed and high quality.