Publications

Export 20 results:
Author Title [ Type(Desc)] Year
Filters: Author is Savchynskyy, B.  [Clear All Filters]
Conference Paper
Kappes, J H, Savchynskyy, B and Schnörr, C (2012). A Bundle Approach To Efficient MAP-Inference by Lagrangian Relaxation. CVPR. Proceedings. 1688-1695
Schmidt, S, Savchynskyy, B, Kappes, J H and Schnörr, C (2011). Evaluation of a First-Order Primal-Dual Algorithm for MRF Energy Minimization. EMMCVPR 2011. Springer. 6819 89-103
Savchynskyy, B and Schmidt, S (2013). Getting Feasible Variable Estimates From Infeasible Ones: MRF Local Polytope Study. Workshop on Inference for Probabilistic Graphical Models at ICCV. Proceedings
Savchynskyy, B, Kappes, J H, Swoboda, P and Schnörr, C (2013). Global MAP-Optimality by Shrinking the Combinatorial Search Area with Convex Relaxation. NIPS. Proceedings. 1950-1958
Swoboda, P, Savchynskyy, B, Kappes, J H and Schnörr, C (2013). Partial Optimality via Iterative Pruning for the Potts Model. Proceedings of the 4th International Conference on Scale Space and Variational Methods in Computer Vision SSVM. 477-488
Swoboda, P, Savchynskyy, B, Kappes, J H and Schnörr, C (2013). Persistency by Pruning for General Graphical Models. submitted to NIPS 2013
Kappes, J, Swoboda, P, Savchynskyy, B, Hazan, T and Schnörr, C (2015). Probabilistic Correlation Clustering and Image Partitioning Using Perturbed Multicuts. Proc. SSVM. Springer
Kappes, J H, Swoboda, P, Savchynskyy, B, Hazan, T and Schnörr, C (2015). Probabilistic Correlation Clustering and Image Partitioning Using Perturbed Multicuts. Proc.~SSVM. SpringerPDF icon Technical Report (1.1 MB)
Savchynskyy, B, Kappes, J H, Schmidt, S and Schnörr, C (2011). A Study of Nesterov's Scheme for Lagrangian Decomposition and MAP Labeling. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR)
Savchynskyy, B, Kappes, J H, Schmidt, S and Schnörr, C (2011). A Study of Nesterov's Scheme for Lagrangian Decomposition and MAP Labeling. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR)PDF icon Technical Report (408.99 KB)
Journal Article
Kappes, J H, Andres, B, Hamprecht, F A, Schnörr, C, Nowozin, S, Batra, D, Kim, S, Kausler, B X, Kröger, T, Lellmann, J, Komodakis, N, Savchynskyy, B and Rother, C (2015). A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems. Int.~J.~Comp.~VisionPDF icon Technical Report (5.12 MB)
Kappes, J H, Andres, B, Hamprecht, F A, Schnörr, C, Nowozin, S, Batra, D, Kim, S, Kausler, B X, Kröger, T, Lellmann, J, Komodakis, N, Savchynskyy, B and Rother, C (2014). A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems. CoRR. http://arxiv.org/abs/1404.0533
Kappes, J H, Andres, B, Hamprecht, F A, Schnörr, C, Nowozin, S, Batra, D, Kim, S, Kausler, B X, Kröger, T, Lellmann, J, Komodakis, N, Savchynskyy, B and Rother, C (2015). A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems. International Journal of Computer Vision. 1-30PDF icon Technical Report (1.5 MB)
Swoboda, P, Kuske, J and Savchynskyy, B (2016). A Dual Ascent Framework for Lagrangean Decomposition of Combinatorial Problems. arXiv, preprint. https://arxiv.org/pdf/1612.05460.pdf
Savchynskyy, B, Schmidt, S, Kappes, J H and Schnörr, C (2012). Efficient MRF Energy Minimization via Adaptive Diminishing Smoothing. UAI. Proceedings. 746-755
Savchynskyy, B and Schmidt, S (2012). Getting Feasible Variable Estimates From Infeasible Ones: MRF Local Polytope Study. arXiv:1210.4081
Kappes, J H, Swoboda, P, Savchynskyy, B, Hazan, T and Schnörr, C (2016). Multicuts and Perturb & MAP for Probabilistic Graph Clustering. J. Math. Imag. Vision. 56 221–237
Swoboda, P, Shekhovtsov, A, Kappes, J H, Schnörr, C and Savchynskyy, B (2016). Partial Optimality by Pruning for MAP-Inference with General Graphical Models. IEEE Trans. Patt. Anal. Mach. Intell. 38 1370–1382
Savchynskyy, B, Kappes, J H, Schmidt, S and Schnörr, C (2011). A Study of Nesterov's Scheme for Lagrangian Decomposition and MAP Labeling. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), accepted as oral presentation. 1817 - 1823