This record contains C++ code and a Docker release for performing curve skeletonization of objects, which have a voxel representation. Curve skeletonization is used to convert a three-dimensional digital object or shape to locally one-dimensional parts; in other words, to reduce the shape information to a more easily processed form. Our algorithm does so for objects whose surface may be noisy, which is a common occurrence when working with data acquired under real-world conditions. This record also includes a test dataset for verifying that the code is running correctly and as examples of how to convert from different file types. This code is a companion to the paper, “Fast and robust curve skeletonization for real-world elongated objects,” by Amy Tabb and Henry Medeiros, published as part of the IEEE WACV 2018 conference. A free, identical version is available on arXiv:1702:07619[cs.CV] (https://arxiv.org/pdf/1702.07619.pdf).
Open source code can be found at:
https://github.com/amy-tabb/CurveSkel-Tabb-Medeiros
Docker container release can be found at:
https://hub.docker.com/r/amytabb/curveskel-tabb-medeiros-docker/
We also include the results of using our code on one of the examples so that you can be sure that the code worked correctly. This folder is named 'A_Result' and can be found in the resource titled 'examples.zip'. A README is included in this record as well as in the examples folder. Both the GitHub and Docker repositories contain the test data to confirm that your data is formatted appropriately.
Problems/Comments/Bugs should be addressed to amy.tabb@ars.usda.gov
Note: Docker link to Curve Skeleton repository resource was added on 7/30/2018
