@conference {Kirillov2017a, title = {InstanceCut: From edges to instances with MultiCut}, booktitle = {Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017}, volume = {2017-Janua}, year = {2017}, pages = {7322{\textendash}7331}, abstract = {This work addresses the task of instance-aware semantic segmentation. Our key motivation is to design a simple method with a new modelling-paradigm, which therefore has a different trade-off between advantages and disadvantages compared to known approaches.Our approach, we term InstanceCut, represents the problem by two output modalities: (i) an instance-agnostic semantic segmentation and (ii) all instance-boundaries. The former is computed from a standard convolutional neural network for semantic segmentation, and the latter is derived from a new instanceaware edge detection model. To reason globally about the optimal partitioning of an image into instances, we combine these two modalities into a novel MultiCut formulation. We evaluate our approach on the challenging CityScapes dataset. Despite the conceptual simplicity of our approach, we achieve the best result among all published methods, and perform particularly well for rare object classes.}, isbn = {9781538604571}, doi = {10.1109/CVPR.2017.774}, author = {Kirillov, Alexander and Levinkov, Evgeny and Bj{\"o}rn Andres and Savchynskyy, Bogdan and Carsten Rother} } @conference {Levinkov2017, title = {Joint graph decomposition \& node labeling: Problem, algorithms, applications}, booktitle = {Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017}, volume = {2017-Janua}, year = {2017}, pages = {1904{\textendash}1912}, abstract = {We state a combinatorial optimization problem whose feasible solutions define both a decomposition and a node labeling of a given graph. This problem offers a common mathematical abstraction of seemingly unrelated computer vision tasks, including instance-separating semantic segmentation, articulated human body pose estimation and multiple object tracking. Conceptually, the problem we state generalizes the unconstrained integer quadratic program and the minimum cost lifted multicut problem, both of which are NP-hard. In order to find feasible solutions efficiently, we define two local search algorithms that converge monotonously to a local optimum, offering a feasible solution at any time. To demonstrate their effectiveness in tackling computer vision tasks, we apply these algorithms to instances of the problem that we construct from published data, using published algorithms. We report state-of-the-art application-specific accuracy for the three above-mentioned applications.}, isbn = {9781538604571}, doi = {10.1109/CVPR.2017.206}, author = {Levinkov, Evgeny and Uhrig, Jonas and Tang, Siyu and Omran, Mohamed and Insafutdinov, Eldar and Kirillov, Alexander and Carsten Rother and Brox, Thomas and Schiele, Bernt and Bj{\"o}rn Andres} } @conference {Royer2016, title = {Convexity shape constraints for image segmentation}, booktitle = {Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition}, volume = {2016-Decem}, year = {2016}, month = {sep}, pages = {402{\textendash}410}, abstract = {Segmenting an image into multiple components is a central task in computer vision. In many practical scenarios, prior knowledge about plausible components is available. Incorporating such prior knowledge into models and algorithms for image segmentation is highly desirable, yet can be non-trivial. In this work, we introduce a new approach that allows, for the first time, to constrain some or all components of a segmentation to have convex shapes. Specifically, we extend the Minimum Cost Multicut Problem by a class of constraints that enforce convexity. To solve instances of this NP-hard integer linear program to optimality, we separate the proposed constraints in the branch-and-cut loop of a state-of-the-art ILP solver. Results on photographs and micrographs demonstrate the effectiveness of the approach as well as its advantages over the state-of-the-art heuristic.}, isbn = {9781467388504}, issn = {10636919}, doi = {10.1109/CVPR.2016.50}, url = {http://arxiv.org/abs/1509.02122}, author = {Royer, Loic A. and Richmond, David L. and Carsten Rother and Bj{\"o}rn Andres and Kainmueller, Dagmar} } @proceedings {6078, title = {An Efficient Fusion Move Algorithm for the Minimum Cost Lifted Multicut Problem}, volume = {LNCS 9906}, year = {2016}, pages = {715-730}, publisher = {Springer}, doi = { 10.1007/978-3-319-46475-6_44}, author = {Thorsten Beier and Bj{\"o}rn Andres and Ullrich K{\"o}the and Fred A. Hamprecht} } @article {Kappes2015b, title = {A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems}, journal = {International Journal of Computer Vision}, volume = {115}, number = {2}, year = {2015}, pages = {155{\textendash}184}, abstract = {Szeliski et al. published an influential study in 2006 on energy minimization methods for Markov random fields. This study provided valuable insights in choosing the best optimization technique for certain classes of problems. While these insights remain generally useful today, the phenomenal success of random field models means that the kinds of inference problems that have to be solved changed significantly. Specifically, the models today often include higher order interactions, flexible connectivity structures, large label-spaces of different cardinalities, or learned energy tables. To reflect these changes, we provide a modernized and enlarged study. We present an empirical comparison of more than 27 state-of-the-art optimization techniques on a corpus of 2453 energy minimization instances from diverse applications in computer vision. To ensure reproducibility, we evaluate all methods in the OpenGM 2 framework and report extensive results regarding runtime and solution quality. Key insights from our study agree with the results of Szeliski et al. for the types of models they studied. However, on new and challenging types of models our findings disagree and suggest that polyhedral methods and integer programming solvers are competitive in terms of runtime and solution quality over a large range of model types.}, keywords = {Benchmark, Combinatorial optimization, Discrete graphical models}, issn = {15731405}, doi = {10.1007/s11263-015-0809-x}, author = {Kappes, J{\"o}rg H and Bj{\"o}rn Andres and Fred A. Hamprecht and Christoph Schn{\"o}rr and Nowozin, Sebastian and Dhruv Batra and Kim, Sungwoong and Kausler, Bernhard X and Kr{\"o}ger, Thorben and Lellmann, Jan and Komodakis, Nikos and Savchynskyy, Bogdan and Carsten Rother} } @article {kappes_15_comparative, title = {A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems}, journal = {International Journal of Computer Vision}, year = {2015}, note = {1}, pages = {1-30}, doi = {10.1007/s11263-015-0809-x}, author = {J{\"o}rg H. Kappes and Bj{\"o}rn Andres and Fred A. Hamprecht and Christoph Schn{\"o}rr and Nowozin, S. and Dhruv Batra and Kim, S. and Bernhard X. Kausler and Thorben Kr{\"o}ger and Lellmann, J. and Komodakis, N. and Savchynskyy, B. and Carsten Rother} } @article {Kappes2015, title = {A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems}, journal = {International Journal of Computer Vision}, volume = {115}, number = {2}, year = {2015}, pages = {155{\textendash}184}, abstract = {Szeliski et al. published an influential study in 2006 on energy minimization methods for Markov random fields. This study provided valuable insights in choosing the best optimization technique for certain classes of problems. While these insights remain generally useful today, the phenomenal success of random field models means that the kinds of inference problems that have to be solved changed significantly. Specifically, the models today often include higher order interactions, flexible connectivity structures, large label-spaces of different cardinalities, or learned energy tables. To reflect these changes, we provide a modernized and enlarged study. We present an empirical comparison of more than 27 state-of-the-art optimization techniques on a corpus of 2453 energy minimization instances from diverse applications in computer vision. To ensure reproducibility, we evaluate all methods in the OpenGM 2 framework and report extensive results regarding runtime and solution quality. Key insights from our study agree with the results of Szeliski et al. for the types of models they studied. However, on new and challenging types of models our findings disagree and suggest that polyhedral methods and integer programming solvers are competitive in terms of runtime and solution quality over a large range of model types.}, keywords = {Benchmark, Combinatorial optimization, Discrete graphical models}, issn = {15731405}, doi = {10.1007/s11263-015-0809-x}, url = {http://hci.iwr.uni-heidelberg.de/opengm2/}, author = {Kappes, J{\"o}rg H and Bj{\"o}rn Andres and Fred A. Hamprecht and Christoph Schn{\"o}rr and Nowozin, Sebastian and Dhruv Batra and Kim, Sungwoong and Kausler, Bernhard X and Kr{\"o}ger, Thorben and Lellmann, Jan and Komodakis, Nikos and Savchynskyy, Bogdan and Carsten Rother} } @article {Kappes2015a, title = {A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems}, journal = {International Journal of Computer Vision}, volume = {115}, number = {2}, year = {2015}, pages = {155{\textendash}184}, abstract = {Szeliski et al. published an influential study in 2006 on energy minimization methods for Markov random fields. This study provided valuable insights in choosing the best optimization technique for certain classes of problems. While these insights remain generally useful today, the phenomenal success of random field models means that the kinds of inference problems that have to be solved changed significantly. Specifically, the models today often include higher order interactions, flexible connectivity structures, large label-spaces of different cardinalities, or learned energy tables. To reflect these changes, we provide a modernized and enlarged study. We present an empirical comparison of more than 27 state-of-the-art optimization techniques on a corpus of 2453 energy minimization instances from diverse applications in computer vision. To ensure reproducibility, we evaluate all methods in the OpenGM 2 framework and report extensive results regarding runtime and solution quality. Key insights from our study agree with the results of Szeliski et al. for the types of models they studied. However, on new and challenging types of models our findings disagree and suggest that polyhedral methods and integer programming solvers are competitive in terms of runtime and solution quality over a large range of model types.}, keywords = {Benchmark, Combinatorial optimization, Discrete graphical models}, isbn = {25164671.25}, issn = {15731405}, doi = {10.1007/s11263-015-0809-x}, author = {Kappes, J{\"o}rg H and Bj{\"o}rn Andres and Fred A. Hamprecht and Christoph Schn{\"o}rr and Nowozin, Sebastian and Dhruv Batra and Kim, Sungwoong and Kausler, Bernhard X and Kr{\"o}ger, Thorben and Lellmann, Jan and Komodakis, Nikos and Savchynskyy, Bogdan and Carsten Rother} } @article {Kappes2015, title = {A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems}, journal = {Int.~J.~Comp.~Vision}, year = {2015}, note = {in press (preprint: arXiv:1404.0533)}, author = {J{\"o}rg H. Kappes and Bj{\"o}rn Andres and Fred A. Hamprecht and Christoph Schn{\"o}rr and Nowozin, S. and Dhruv Batra and Kim, S. and Bernhard X. Kausler and Thorben Kr{\"o}ger and Lellmann, J. and Komodakis, N. and Savchynskyy, B. and Carsten Rother} } @article {kappes_14_comparative, title = {A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems}, journal = {CoRR}, year = {2014}, note = {1}, url = {http://arxiv.org/abs/1404.0533}, author = {J{\"o}rg H. Kappes and Bj{\"o}rn Andres and Fred A. Hamprecht and Christoph Schn{\"o}rr and Nowozin, S. and Dhruv Batra and Kim, S. and Bernhard X. Kausler and Thorben Kr{\"o}ger and Lellmann, J. and Komodakis, N. and Savchynskyy, B. and Carsten Rother} } @article {kappes-1014-bench-arxiv, title = {A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems}, journal = {CoRR}, volume = {abs/1404.0533}, year = {2014}, url = {http://hci.iwr.uni-heidelberg.de/opengm2/}, author = {J{\"o}rg H. Kappes and Bj{\"o}rn Andres and Fred A. Hamprecht and Christoph Schn{\"o}rr and Nowozin, Sebastian and Dhruv Batra and Kim, Sungwoong and Bernhard X. Kausler and Thorben Kr{\"o}ger and Lellmann, Jan and Komodakis, Nikos and Savchynskyy, Bogdan and Carsten Rother} } @conference {kappes_13_comparative, title = {A Comparative Study of Modern Inference Techniques for Discrete Energy Minimization Problems}, booktitle = {CVPR 2013. Proceedings}, year = {2013}, note = {1}, doi = {10.1109/CVPR.2013.175}, author = {J{\"o}rg H. Kappes and Bj{\"o}rn Andres and Fred A. Hamprecht and Christoph Schn{\"o}rr and Nowozin, S. and Dhruv Batra and Sungwoong, K. and Bernhard X. Kausler and Lellmann, J. and Komodakis, N. and Carsten Rother} } @conference {Kappes-2013-benchmark, title = {A Comparative Study of Modern Inference Techniques for Discrete Energy Minimization Problem}, booktitle = {CVPR}, year = {2013}, author = {J{\"o}rg H. Kappes and Bj{\"o}rn Andres and Fred A. Hamprecht and Christoph Schn{\"o}rr and Nowozin, Sebastian and Dhruv Batra and Kim, Sungwoong and Bernhard X. Kausler and Lellmann, Jan and Komodakis, Nikos and Carsten Rother} } @article {andres_11_3d, title = {3D Segmentation of SBFSEM Images of Neuropil by a Graphical Model over Supervoxel Boundaries}, journal = {Medical Image Analysis}, volume = {16 (2012)}, year = {2012}, note = {1}, pages = {796-805}, doi = {10.1016/j.media.2011.11.004}, author = {Bj{\"o}rn Andres and Ullrich K{\"o}the and Thorben Kr{\"o}ger and Helmstaedter, M. and K. L. Briggmann and Denk, W. and Fred A. Hamprecht} } @conference {kausler_12_discrete, title = {A Discrete Chain Graph Model for 3d+t Cell Tracking with High Misdetection Robustness}, booktitle = {ECCV 2012. Proceedings}, volume = {7574}, year = {2012}, note = {1}, pages = {144-157}, doi = {10.1007/978-3-642-33712-3_11}, author = {Bernhard X. Kausler and Schiegg, M. and Bj{\"o}rn Andres and Lindner, M. and Ullrich K{\"o}the and Leitte, H. and Wittbrodt, J. and Hufnagel, L. and Fred A. Hamprecht} } @article {funke_12_efficient, title = {Efficient Automatic 3D-Reconstruction of Branching Neurons from EM Data}, journal = {CVPR 2012. Proceedings}, year = {2012}, note = {1}, pages = {1004-1011}, doi = {10.1109/CVPR.2012.6247777}, author = {Funke, J. and Bj{\"o}rn Andres and Fred A. Hamprecht and A. Cardona and Cook, M.} } @conference {andres_12_globally, title = {Globally Optimal Closed-Surface Segmentation for Connectomics}, booktitle = {ECCV 2012. Proceedings, Part 3}, number = {7574}, year = {2012}, note = {1}, pages = {778-791}, doi = {10.1007/978-3-642-33712-3_56}, author = {Bj{\"o}rn Andres and Thorben Kr{\"o}ger and K. L. Briggmann and Denk, W. and Norogod, N. and G. W. Knott and Ullrich K{\"o}the and Fred A. Hamprecht} } @conference {Andres12, title = {The Lazy Flipper: Efficient Depth-limited Exhaustive Search in Discrete Graphical Models}, booktitle = {ECCV 2012}, year = {2012}, author = {Bj{\"o}rn Andres and J{\"o}rg H. Kappes and Thorsten Beier and Ullrich K{\"o}the and Fred A. Hamprecht} } @conference {Andres12, title = {The Lazy Flipper: Efficient Depth-limited Exhaustive Search in Discrete Graphical Models}, booktitle = {ECCV 2012}, year = {2012}, author = {Bj{\"o}rn Andres and J{\"o}rg Hendrik Kappes and Thorsten Beier and Ullrich K{\"o}the and Fred A. Hamprecht} } @conference {andres_12_lazy, title = {The Lazy Flipper: Efficient Depth-Limited Exhaustive Search in Discrete Graphical Models}, booktitle = {Computer Vision - {ECCV} 2012 - 12th European Conference on Computer Vision, Florence, Italy, October 7-13, 2012, Proceedings, Part {VII}}, year = {2012}, doi = {10.1007/978-3-642-33786-4_12}, url = {http://dx.doi.org/10.1007/978-3-642-33786-4_12}, author = {Bj{\"o}rn Andres and J{\"o}rg H. Kappes and Thorsten Beier and Ullrich K{\"o}the and Fred A. Hamprecht} } @article {andres2012opengm, title = {OpenGM: A C++ Library for Discrete Graphical Models}, journal = {ArXiv e-prints}, year = {2012}, note = {Projectpage: http://hci.iwr.uni-heidelberg.de/opengm2/}, author = {Bj{\"o}rn Andres and Thorsten Beier and Kappes, J{\"o}rg H.} } @phdthesis {andres_11_automated, title = {Automated Segmentation of Large 3D Images of Nervous Systems Using a Higher-order Graphical Model}, year = {2011}, note = {1}, publisher = {University of Heidelberg}, author = {Bj{\"o}rn Andres} } @conference {Kappes11, title = {Globally Optimal Image Partitioning by Multicuts}, booktitle = {EMMCVPR}, year = {2011}, publisher = {Springer}, organization = {Springer}, author = {J{\"o}rg H. Kappes and Speth, Markus and Bj{\"o}rn Andres and Reinelt, Gerhard and Christoph Schn{\"o}rr} } @conference {Kappes11, title = {Globally Optimal Image Partitioning by Multicuts}, booktitle = {EMMCVPR}, year = {2011}, publisher = {Springer}, organization = {Springer}, author = {J{\"o}rg Hendrik Kappes and Markus Speth and Bj{\"o}rn Andres and Gerhard Reinelt and Christoph Schn{\"o}rr} } @conference {kappes_11_globally, title = {Globally Optimal Image Partitioning by Multicuts}, booktitle = {EMMCVPR}, year = {2011}, note = {1}, pages = {31-44}, publisher = {Springer}, organization = {Springer}, doi = {10.1007/978-3-642-23094-3_3}, author = {J{\"o}rg H. Kappes and Speth, M. and Bj{\"o}rn Andres and Reinelt, G. and Christoph Schn{\"o}rr} } @conference {Andres11, title = {Probabilistic Image Segmentation with Closedness Constraints}, booktitle = {Proceedings of ICCV}, year = {2011}, author = {Bj{\"o}rn Andres and J{\"o}rg H. Kappes and Thorsten Beier and Ullrich K{\"o}the and Fred A. Hamprecht} } @conference {Andres11, title = {Probabilistic Image Segmentation with Closedness Constraints}, booktitle = {Proceedings of ICCV}, year = {2011}, author = {Bj{\"o}rn Andres and J{\"o}rg H. Kappes and Thorsten Beier and Ullrich K{\"o}the and Fred A. Hamprecht} } @conference {andres_11_probabilistic, title = {Probabilistic Image Segmentation with Closedness Constraints}, booktitle = {ICCV, Proceedings}, year = {2011}, note = {1}, pages = {2611 - 2618}, doi = {10.1109/ICCV.2011.6126550}, author = {Bj{\"o}rn Andres and J{\"o}rg H. Kappes and Thorsten Beier and Ullrich K{\"o}the and Fred A. Hamprecht} } @conference {Kappes-DAGM2010, title = {An Empirical Comparison of Inference Algorithms for Graphical Models with Higher Order Factors Using OpenGM}, booktitle = {Pattern Recognition, Proc.~32th DAGM Symposium}, year = {2010}, author = {Bj{\"o}rn Andres and J{\"o}rg H. Kappes and Ullrich K{\"o}the and Christoph Schn{\"o}rr and Fred A. Hamprecht} } @conference {andres_10_empirical, title = {An Empirical Comparison of Inference Algorithms for Graphical Models with Higher Order Factors Using OpenGM}, booktitle = {Pattern Recognition, Proc.~32th DAGM Symposium}, number = {6376}, year = {2010}, note = {1}, pages = {353-362}, doi = {10.1007/978-3-642-15986-2_36}, author = {Bj{\"o}rn Andres and J{\"o}rg H. Kappes and Ullrich K{\"o}the and Christoph Schn{\"o}rr and Fred A. Hamprecht} } @conference {koethe_10_geometric, title = {Geometric Analysis of 3D Electron Microscopy Data}, booktitle = {Proceedings of Workshop on Discrete Geometry and Mathematical Morphology (WADGMM)}, year = {2010}, note = {1}, pages = {22-26}, author = {Ullrich K{\"o}the and Bj{\"o}rn Andres and Thorben Kr{\"o}ger and Fred A. Hamprecht}, editor = {Ullrich K{\"o}the and Montanvert, A. and Soille, P.} } @article {andres_10_how, title = {How to Extract the Geometry and Topology from Very Large 3D Segmentations}, journal = {ArXiv e-prints}, year = {2010}, note = {1}, url = {http://arxiv.org/abs/1009.6215}, author = {Bj{\"o}rn Andres and Ullrich K{\"o}the and Thorben Kr{\"o}ger and Fred A. Hamprecht} } @article {andres_10_lazy, title = {The Lazy Flipper: MAP Inference in Higher-Order Graphical Models by Depth-limited Exhaustive Search}, journal = {ArXiv e-prints}, year = {2010}, note = {1}, url = {http://arxiv.org/abs/1009.4102}, author = {Bj{\"o}rn Andres and J{\"o}rg H. Kappes and Ullrich K{\"o}the and Fred A. Hamprecht} } @article {andres_10_runtime, title = {Runtime-Flexible Multi-dimensional Views and Arrays for C++98 and C++0x}, journal = {ArXiv e-prints}, year = {2010}, note = {1}, url = {http://arxiv.org/abs/1008.2909v1}, author = {Bj{\"o}rn Andres and Ullrich K{\"o}the and Thorben Kr{\"o}ger and Fred A. Hamprecht} } @conference {andres_09_quantitative, title = {Quantitative Assessment of Image Segmentation Quality by Random Walk Relaxation Times}, booktitle = {Pattern Recognition. 31st DAGM Symposium, Jena, Germany, September 9-11, 2009. Proceedings}, volume = {5748}, year = {2009}, note = {1}, pages = {502-511}, publisher = {Springer}, organization = {Springer}, doi = {10.1007/978-3-642-03798-6}, author = {Bj{\"o}rn Andres and Ullrich K{\"o}the and Bonea, A. and Nadler, B. and Fred A. Hamprecht} } @conference {andres2008, title = {On errors-in-variables regression with arbitrary covariance and its application to optical flow estimation}, booktitle = {IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2008}, year = {2008}, pages = {1--6}, publisher = {IEEE}, organization = {IEEE}, abstract = {Linear inverse problems in computer vision, including motion estimation, shape fitting and image reconstruction, give rise to parameter estimation problems with highly correlated errors in variables. Established total least squares methods estimate the most likely corrections Acirc and bcirc to a given data matrix [A, b] perturbed by additive Gaussian noise, such that there exists a solution y with [A + Acirc, b +bcirc]y = 0. In practice, regression imposes a more restrictive constraint namely the existence of a solution x with [A + Acirc]x = [b + bcirc]. In addition, more complicated correlations arise canonically from the use of linear filters. We, therefore, propose a maximum likelihood estimator for regression in the general case of arbitrary positive definite covariance matrices. We show that Acirc, bcirc and x can be found simultaneously by the unconstrained minimization of a multivariate polynomial which can, in principle, be carried out by means of a Grobner basis. Results for plane fitting and optical flow computation indicate the superiority of the proposed method.}, doi = {10.1109/CVPR.2008.4587571}, author = {Bj{\"o}rn Andres and Claudia Kondermann and Daniel Kondermann and Fred A. Hamprecht and Christoph S. Garbe} } @article {andres_08_errors-in-variables, title = {On errors-in-variables regression with arbitrary covariance and its application to optical flow estimation}, journal = {Computer Vision and Pattern Recognition, 2008. CVPR 2008. IEEE Conference on}, year = {2008}, note = {1}, pages = {1-6}, doi = {10.1109/CVPR.2008.4587571}, author = {Bj{\"o}rn Andres and Claudia Kondermann and Daniel Kondermann and Ullrich K{\"o}the and Fred A. Hamprecht and Christoph S. Garbe} } @incollection {garbe_08_nonlinear, title = {Nonlinear Analysis of Multi-Dimensional Signals}, year = {2008}, note = {1}, pages = {231-288}, publisher = {Springer}, doi = {10.1007/978-3-540-75632-3_7}, author = {Christoph S. Garbe and Kai Krajsek and Pavel Pavlov and Bj{\"o}rn Andres and Matthias M{\"u}hlich and Ingo Stuke and Cicero Mota and Martin B{\"o}hme and Martin Haker and Schuchert, T. and Hanno Scharr and Til Aach and Erhardt Barth}, editor = {R. Dahlhaus and J. Kurths and Timmer, J. and Peter Maass} } @incollection {garbe2008, title = {Nonlinear analysis of multi-dimensional signals: local adaptive estimation of complex motion and orientation patterns}, year = {2008}, pages = {231-288}, publisher = {Springer}, abstract = {We consider the general task of accurately detecting and quantifying orientations in n-dimensional signals s. The main emphasis will be placed on the estimation of motion, which can be thought of as orientation in spatiotemporal signals. Associated problems such as the optimization of matched kernels for deriving isotropic and highly accurate gradients from the signals, optimal integration of local models, and local model selection will also be addressed.}, doi = {10.1007/978-3-540-75632-3_7}, author = {Christoph S. Garbe and Kai Krajsek and Pavel Pavlov and Bj{\"o}rn Andres and Matthias M{\"u}hlich and Ingo Stuke and Cicero Mota and Martin B{\"o}hme and Martin Haker and Tobias Schucher and Hanno Scharr and Til Aach and Erhardt Barth}, editor = {R. Dahlhaus and J. Kurths and Timmer, J. and Peter Maass} } @conference {andres_08_segmentation, title = {Segmentation of SBFSEM Volume Data of Neural Tissue by Hierarchical Classification}, booktitle = {Pattern Recognition. 30th DAGM Symposium Munich, Germany, June 10-13, 2008. Proceedings}, volume = {5096}, year = {2008}, note = {1}, pages = {142-152}, publisher = {Springer}, organization = {Springer}, doi = {10.1007/978-3-540-69321-5_15}, author = {Bj{\"o}rn Andres and Ullrich K{\"o}the and Helmstaedter, M. and Denk, W. and Fred A. Hamprecht}, editor = {Gerhard Rigoll} } @mastersthesis {andres2007a, title = {Model Selection in Optical Flow-Based Motion Estimation by Means of Residual Analysis}, year = {2007}, school = {University of Heidelberg}, author = {Bj{\"o}rn Andres} } @mastersthesis {andres_07_model, title = {Model Selection in Optical Flow-Based Motion Estimation by Means of Residual Analysis}, year = {2007}, note = {1}, school = {University of Heidelberg}, author = {Bj{\"o}rn Andres} } @conference {andres_07_selection, title = {Selection of Local Optical Flow Models by Means of Residual Analysis}, booktitle = {Pattern Recognition}, volume = {4713}, year = {2007}, note = {1}, pages = {72-81}, publisher = {Springer}, organization = {Springer}, doi = {10.1007/978-3-540-74936-3_8}, author = {Bj{\"o}rn Andres and Fred A. Hamprecht and Christoph S. Garbe}, editor = {Christoph Schn{\"o}rr and Bernd J{\"a}hne and Fred A. Hamprecht} } @conference {andres2007, title = {Selection of local optical flow models by means of residual analysis}, booktitle = {Proceedings of the 29th DAGM Symposium on Pattern Recognition}, year = {2007}, pages = {72--81}, publisher = {Springer}, organization = {Springer}, abstract = {This contribution presents a novel approach to the challenging problem of model selection in motion estimation from sequences of images. New light is cast on parametric models of local optical flow. These models give rise to parameter estimation problems with highly correlated errors in variables (EIV). Regression is hence performed by equilibrated total least squares. The authors suggest to adaptively select motion models by testing local empirical regression residuals to be in accordance with the probability distribution that is theoretically predicted by the EIV model. Motion estimation with residual-based model selection is examined on artificial sequences designed to test specifically for the properties of the model selection process. These simulations indicate a good performance in the exclusion of inappropriate models and yield promising results in model complexity control.}, doi = {10.1007/978-3-540-74936-3_8}, author = {Bj{\"o}rn Andres and Christoph S. Garbe and Christoph Schn{\"o}rr and Bernd J{\"a}hne}, editor = {Fred A. Hamprecht and Fred A. Hamprecht} }