A fast projection method for connectivity constraints in image segmentation
Stühmer J, Cremers D (2015)
Publication Type: Conference contribution
Publication year: 2015
Journal
Publisher: Springer Verlag
Book Volume: 8932
Pages Range: 183-196
Conference Proceedings Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Event location: Hong Kong, CHN
ISBN: 9783319146119
DOI: 10.1007/978-3-319-14612-6_14
Abstract
We propose to solve an image segmentation problem with connectivity constraints via projection onto the constraint set. The constraints form a convex set and the convex image segmentation problem with a total variation regularizer can be solved to global optimality in a primal-dual framework. Efficiency is achieved by directly computing the update of the primal variable via a projection onto the constraint set, which results in a special quadratic programming problem similar to the problems studied as isotonic regression methods in statistics, which can be solved with O(n log n) complexity. We show that especially for segmentation problems with long range connections this method is by orders of magnitudes more efficient, both in iteration number and runtime, than solving the dual of the constrained optimization problem. Experiments validate the usefulness of connectivity constraints for segmenting thin structures such as veins and arteries in medical image analysis.
Involved external institutions
Germany (DE)How to cite
APA:
Stühmer, J., & Cremers, D. (2015). A fast projection method for connectivity constraints in image segmentation. In Xue-Cheng Tai, Egil Bae, Tony F. Chan, Marius Lysaker (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 183-196). Hong Kong, CHN: Springer Verlag.
MLA:
Stühmer, Jan, and Daniel Cremers. "A fast projection method for connectivity constraints in image segmentation." Proceedings of the 10th International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition, EMMCVPR 2015, Hong Kong, CHN Ed. Xue-Cheng Tai, Egil Bae, Tony F. Chan, Marius Lysaker, Springer Verlag, 2015. 183-196.
BibTeX: Download