233 Hits in 7.0 sec

Reconstructing Curvilinear Networks Using Path Classifiers and Integer Programming

Engin Turetken, Fethallah Benmansour, Bjoern Andres, Przemyslaw Glowacki, Hanspeter Pfister, Pascal Fua
2016 IEEE Transactions on Pattern Analysis and Machine Intelligence  
By representing the image data as a graph of potential paths, we first show how to weight these paths using discriminatively-trained classifiers that are both robust and generic enough to be applied to  ...  We then present an Integer Programming approach to finding the optimal subset of paths, subject to structural and topological constraints that eliminate implausible solutions.  ...  insights with us.  ... 
doi:10.1109/tpami.2016.2519025 pmid:26891482 fatcat:ss7fc7jpizasrcnr3f5epcg3iq

Stochastic model for curvilinear structure reconstruction using morphological profiles

Seong-Gyun Jeong, Yuliya Tarabalka, Josiane Zerubia
2015 2015 IEEE International Conference on Image Processing (ICIP)  
In this work, we propose a stochastic model for curvilinear structure reconstruction using morphological profiles of path opening operator.  ...  We apply the support vector machine classifier to obtain initial probabilities to belong to line network for each pixel.  ...  We proposed a new curvilinear structure reconstruction algorithm which exploits morphological profiles of path opening operator.  ... 
doi:10.1109/icip.2015.7351470 dblp:conf/icip/JeongTZ15 fatcat:qprp2v6i3veiro6jhgibscxylq

Reconstructing Loopy Curvilinear Structures Using Integer Programming

Engin Turetken, Fethallah Benmansour, Bjoern Andres, Hanspeter Pfister, Pascal Fua
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
At the heart of our method is an Integer Programming formulation that allows us to find the global optimum of an objective function designed to allow cycles but penalize spurious junctions and early terminations  ...  We propose a novel approach to automated delineation of linear structures that form complex and potentially loopy networks.  ...  The MAP formulation [25] addresses these issues by using path classifiers to score the paths and introducing a Mixed Integer Programming approach to guaranteeing optimality of the resulting solution.  ... 
doi:10.1109/cvpr.2013.238 dblp:conf/cvpr/TuretkenBAPF13 fatcat:p465dceyqbeyxelqduedj7tfry

Active Learning and Proofreading for Delineation of Curvilinear Structures [chapter]

Agata Mosinska, Jakub Tarnawski, Pascal Fua
2017 Lecture Notes in Computer Science  
We demonstrate the effectiveness of our approach on microscopy images depicting blood vessels and neurons.  ...  In an Active Learning context, we identify parts of linear structures that should be annotated first in order to train a classifier effectively.  ...  Introduction Complex and extensive curvilinear structures include blood vessels, pulmonary bronchi, nerve fibers and neuronal networks among others.  ... 
doi:10.1007/978-3-319-66185-8_19 fatcat:yjwzbwa5izgf5ork5h42tjf7hy

Automated reconstruction of tree structures using path classifiers and Mixed Integer Programming

E. Turetken, F. Benmansour, P. Fua
2012 2012 IEEE Conference on Computer Vision and Pattern Recognition  
In this paper, we formulate the delineation problem as one of solving a Quadratic Mixed Integer Program (Q-MIP) in a graph of potential paths, which can be done optimally up to a very small tolerance.  ...  They are used to find the most likely paths using a search mechanism that can be either local or global.  ...  We use these weights and solve a Mixed Integer Program to compute the maximum-likelihood directed tree in this graph (see Fig. 2 (e)).  ... 
doi:10.1109/cvpr.2012.6247722 dblp:conf/cvpr/TuretkenBF12 fatcat:nme4irif4nb5pewlodxxrrvh6e

Reconstructing Evolving Tree Structures in Time Lapse Sequences

Przemyslaw Glowacki, Miguel Amavel Pinheiro, Engin Turetken, Raphael Sznitman, Daniel Lebrecht, Jan Kybic, Anthony Holtmaat, Pascal Fua
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
We show that this problem can be formulated as a Quadratic Mixed Integer Program and solved efficiently.  ...  We propose an approach to reconstructing tree structures that evolve over time in 2D images and 3D image stacks such as neuronal axons or plant branches.  ...  This last step can be done using Minimum Spanning Trees (MST) [6, 26, 28] , Shortest Path Trees (SPT) [19] , k-Minimum Spanning Trees (k-MST) [25] , and Quadratic Mixed Integer Programming (QMIP) [  ... 
doi:10.1109/cvpr.2014.388 dblp:conf/cvpr/GlowackiPTSLKHF14 fatcat:o2ammsv7zrcfpgodfqsregiiju

Joint Segmentation and Path Classification of Curvilinear Structures [article]

Agata Mosinska, Mateusz Kozinski, Pascal Fua
2019 arXiv   pre-print
Most existing delineation approaches first perform binary segmentation of the image and then refine it using either a set of hand-designed heuristics or a separate classifier that assigns likelihood to  ...  In our work, we bridge the gap between segmentation and path classification by training a deep network that performs those two tasks simultaneously.  ...  It also performs Mixed-Integer Programming optimization to find the optimal subgraph.  ... 
arXiv:1905.03892v1 fatcat:wx6t5ibn6zgu7httul327oc5om

Reconstructing Evolving Tree Structures in Time Lapse Sequences by Enforcing Time-Consistency

Przemyslaw Glowacki, Miguel Amavel Pinheiro, Agata Mosinska, Engin Turetken, Daniel Lebrecht, Raphael Sznitman, Anthony Holtmaat, Jan Kybic, Pascal Fua
2018 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We formulate the problem as a Quadratic Mixed Integer Program and demonstrate the additional robustness that comes from using all available visual clues at once, instead of working frame by frame.  ...  We propose a novel approach to reconstructing curvilinear tree structures evolving over time, such as road networks in 2D aerial images or neural structures in 3D microscopy stacks acquired in vivo.  ...  As in [5] , the final set of trees can then be reconstructed by solving a Quadratic Mixed Integer Program (QMIP) problem.  ... 
doi:10.1109/tpami.2017.2680444 pmid:28333621 fatcat:ui44zruzuvdf3ihew65f7nlnym

Modeling brain circuitry over a wide range of scales

Pascal Fua, Graham W. Knott
2015 Frontiers in Neuroanatomy  
model the 3D structure of the dendritic network.  ...  They include delineating dendritic arbors from LM imagery, segmenting organelles from EM, and combining the two into a consistent representation.  ...  one of solving an Integer Program (IP) on a graph of potential tubular paths.  ... 
doi:10.3389/fnana.2015.00042 pmid:25904852 pmcid:PMC4387921 fatcat:yfafumi5tveq7mpt33yalynqvm

Reconstructing cerebrovascular networks under local physiological constraints by integer programming

Markus Rempfler, Matthias Schneider, Giovanna D. Ielacqua, Xianghui Xiao, Stuart R. Stock, Jan Klohs, Gábor Székely, Bjoern Andres, Bjoern H. Menze
2015 Medical Image Analysis  
The method accounts for both image evidence and geometric relationships between vessels by solving an integer program, which is shown to yield the maximum a posteriori (MAP) estimate to a probabilistic  ...  We finally discuss properties of the networks obtained under different tracking and pruning approaches.  ...  [20] introduced recently an integer programming 35 approach that evaluates path coherence and connectivity of general curvilinear structures, such as streets in remote-sensing images or vessels in  ... 
doi:10.1016/ pmid:25977158 fatcat:lnzz5dd7nfh2vhdldznuzbjbjm

Scanning the Issue

Petros Ioannou, A. V. Bal Balakrishnan
2018 IEEE transactions on intelligent transportation systems (Print)  
Analysis from the aspects of productivity, topics, citations, usage, and coauthorship networks is presented.  ...  The results show that China is playing a critical role in ITS research during this period but the interinstitution collaborations are less prevalent than it used to be.  ...  In the first stage of the proposed algorithm, for a static environment, deterministic robust solutions are determined based on the mixed integer linear programming.  ... 
doi:10.1109/tits.2018.2866887 fatcat:3brvnbxecnegbdwnf5f56cg4ey

Trajectory Grouping with Curvature Regularization for Tubular Structure Tracking [article]

Li Liu, Da Chen, Minglei Shu, Baosheng Li, Huazhong Shu, Michel Paques, Laurent D. Cohen
2021 arXiv   pre-print
The proposed approach can benefit from the local smoothness prior on tubular structures and the global optimality of the used graph-based path searching scheme.  ...  Basically, we take into account the prescribed tubular trajectories and curvature-penalized geodesic paths to seek suitable shortest paths.  ...  Fua, “Reconstructing curvilinear networks using path classifiers and vision and image analysis, such as minimal geodesic paths, active contours, integer programming,” IEEE Trans.  ... 
arXiv:2003.03710v4 fatcat:ro4fekyvh5cpfpzn6imbealx5q

14th International Symposium on Mathematical Programming

1990 Mathematical programming  
Finally I report numerical results of a comparison between the variant and a minimization approach using penalty functions.  ...  If we use a decomposition approach in order to solve a minimization problem we often get an objective function in such a w a y that its domain dom 6 = n is not given explicitely to us.  ...  Using a combination of optimization methodologies integer programming, dynamic multicommodity network ows and decomposition techniques, we solve large scale realistic size problems fast.  ... 
doi:10.1007/bf01580875 fatcat:3jtclwmntzgjxkqs5uecombdaa

Simultaneous segmentation and anatomical labeling of the cerebral vasculature

David Robben, Engin Türetken, Stefan Sunaert, Vincent Thijs, Guy Wilms, Pascal Fua, Frederik Maes, Paul Suetens
2016 Medical Image Analysis  
We formulate the latter problem as an Integer Program (IP), which can be solved optimally using a branch-and-cut algorithm.  ...  knowledge about the geometry and connectivity of the vasculature.  ...  The MRA brain images from healthy volunteers used in this paper were collected and made available by the CASILab at The University of North Carolina at Chapel Hill and were distributed by the MI-DAS Data  ... 
doi:10.1016/ pmid:27131026 fatcat:2l6jkmzilvaklpcge3l2q44xgq

tfShearlab: The TensorFlow Digital Shearlet Transform for Deep Learning [article]

Héctor Andrade-Loarca, Gitta Kutyniok
2020 arXiv   pre-print
primal-dual reconstruction method for low dose computed tomography in running time.  ...  As such it has been shown to be well suited for image preprocessing in combination with data-driven methods such as deep neural networks.  ...  This network was able to classify small images of handwritten digits with high accuracy.  ... 
arXiv:2006.04591v1 fatcat:rur4ghxctvdwtljtixua4nvuxm
« Previous Showing results 1 — 15 out of 233 results