ML Anthology

Discrete-Continuous ADMM for Transductive Inference in Higher-Order MRFs

Emanuel Laude, Jan-Hendrik Lange, Jonas Schüpfer, Csaba Domokos, Laura Leal-Taixé, Frank R. Schmidt, Bjoern Andres, Daniel Cremers
CVPR 2018
doi:10.1109/CVPR.2018.00174 /cvpr/2018/laude2018cvpr-discretecontinuous/

Abstract

This paper introduces a novel algorithm for transductive inference in higher-order MRFs, where the unary energies are parameterized by a variable classifier. The considered task is posed as a joint optimization problem in the continuous classifier parameters and the discrete label variables. In contrast to prior approaches such as convex relaxations, we propose an advantageous decoupling of the objective function into discrete and continuous subproblems and a novel, efficient optimization method related to ADMM. This approach preserves integrality of the discrete label variables and guarantees global convergence to a critical point. We demonstrate the advantages of our approach in several experiments including video object segmentation on the DAVIS data set and interactive image segmentation.

PDF CVPR Semantic Scholar

Cite

Text

Laude et al. "Discrete-Continuous ADMM for Transductive Inference in Higher-Order MRFs." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018. doi:10.1109/CVPR.2018.00174

Markdown

[Laude et al. "Discrete-Continuous ADMM for Transductive Inference in Higher-Order MRFs." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018.](https://mlanthology.org/cvpr/2018/laude2018cvpr-discretecontinuous/) doi:10.1109/CVPR.2018.00174

BibTeX

@inproceedings{laude2018cvpr-discretecontinuous,
  title     = {{Discrete-Continuous ADMM for Transductive Inference in Higher-Order MRFs}},
  author    = {Laude, Emanuel and Lange, Jan-Hendrik and Schüpfer, Jonas and Domokos, Csaba and Leal-Taixé, Laura and Schmidt, Frank R. and Andres, Bjoern and Cremers, Daniel},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2018},
  doi       = {10.1109/CVPR.2018.00174},
  url       = {https://mlanthology.org/cvpr/2018/laude2018cvpr-discretecontinuous/}
}