Filters








14,677 Hits in 3.4 sec

Convexity Shape Prior for Binary Segmentation

Lena Gorelick, Olga Veksler, Yuri Boykov, Claudia Nieuwenhuis
2017 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Convexity is a known important cue in human vision. We propose shape convexity as a new high-order regularization constraint for binary image segmentation.  ...  Our experiments demonstrate general usefulness of the proposed convexity shape prior on synthetic and real image segmentation examples.  ...  ACKNOWLEDGEMENTS We are thankful for generous support by Canadian NSERC Discovery and RTI Programs  ... 
doi:10.1109/tpami.2016.2547399 pmid:28103187 fatcat:jqrraaf2ojhyhcmvx76xs3n2um

Weakly Convex Coupling Continuous Cuts and Shape Priors [chapter]

Bernhard Schmitzer, Christoph Schnörr
2012 Lecture Notes in Computer Science  
Specifically, we combine total variation based continuous cuts for image segmentation and convex relaxations of Markov Random Field based shape priors learned from shape databases.  ...  We introduce a novel approach to variational image segmentation with shape priors.  ...  level set based functionals have been used for image segmentation; adding a shape penalty functional compromises convexity of the overall approach.  ... 
doi:10.1007/978-3-642-24785-9_36 fatcat:saetiq6d6jdphc4m4u5guopygu

Convex Shape Representation with Binary Labels for Image Segmentation: Models and Fast Algorithms [article]

Shousheng Luo and Xue-Cheng Tai and Yang Wang
2020 arXiv   pre-print
We present a novel and effective binary representation for convex shapes. We show the equivalence between the shape convexity and some properties of the associated indicator function.  ...  In order to show the effectiveness of the proposed representation approach, we incorporate it with a probability based model for object segmentation with convexity prior.  ...  Image segmentation model with convex prior In this section we will incorporate the proposed binary representation with a segmentation model for convex object segmentation.  ... 
arXiv:2002.09600v1 fatcat:gx32632uhjhgfdvp3mdvmvhbdm

K-convexity Shape Priors for Segmentation [chapter]

Hossam Isack, Lena Gorelick, Karin Ng, Olga Veksler, Yuri Boykov
2018 Lecture Notes in Computer Science  
For example, one approach segments an object via optimizing its k coverage by disjoint convex parts, which we show is highly sensitive to local minima.  ...  Since an arbitrary shape can always be divided into convex parts, our regularization model restricts the number of such parts. Previous k-part shape priors are limited to disjoint parts.  ...  Menda for providing the liver data NIH grant U01-CA140206. This work was also supported by NSERC Discovery and RTI grants (Canada) for Y. Boykov and O. Veksler.  ... 
doi:10.1007/978-3-030-01252-6_3 fatcat:327s7kwvfvgl3gxhhruyn2zjha

Convex Shape Prior for Deep Neural Convolution Network based Eye Fundus Images Segmentation [article]

Jun Liu, Xue-Cheng Tai, Shousheng Luo
2020 arXiv   pre-print
Convex Shapes (CS) are common priors for optic disc and cup segmentation in eye fundus images. It is important to design proper techniques to represent convex shapes.  ...  In the dual space, the convex shape prior can be guaranteed by a simple quadratic constraint on a binary representation of the shapes.  ...  For convex shape prior with binary segmentation, a quadratic convex shape constraint is proposed in [15] with discrete curvature κ 0 derived from thresholding dynamics (TD).  ... 
arXiv:2005.07476v1 fatcat:sxxdooobefhr5ikvpn6yf3e3tu

Convexity Shape Prior for Segmentation [chapter]

Lena Gorelick, Olga Veksler, Yuri Boykov, Claudia Nieuwenhuis
2014 Lecture Notes in Computer Science  
Convexity is known as an important cue in human vision. We propose shape convexity as a new high-order regularization constraint for binary image segmentation.  ...  Unlike standard secondorder length regularization, our convexity prior is scale invariant, does not have shrinking bias, and is virtually parameter-free.  ...  (d) segmentation with convexity shape prior.  ... 
doi:10.1007/978-3-319-10602-1_44 fatcat:5sxj2g5oszeatgedq5jobpa5na

Hedgehog Shape Priors for Multi-Object Segmentation

Hossam Isack, Olga Veksler, Milan Sonka, Yuri Boykov
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Star-convexity prior is popular for interactive single object segmentation due to its simplicity and amenability to binary graph cut optimization.  ...  We propose a more general multi-object segmentation approach. Moreover, each object can be constrained by a more descriptive shape prior, "hedgehog".  ...  Sue O'Dorisio and Yusuf Menda for providing the PET-CT liver data (NIH grant U01-CA140206). This work was also supported by NSERC Discovery and RTI grants (Canada) for Y. Boykov and O. Veksler.  ... 
doi:10.1109/cvpr.2016.267 dblp:conf/cvpr/IsackVSB16 fatcat:wh5dasbv4va2pnphakfho4ydqm

A convex framework for image segmentation with moment constraints

Maria Klodt, Daniel Cremers
2011 2011 International Conference on Computer Vision  
In this paper, we will show that shape priors in terms of moment constraints can be imposed within the convex optimization framework, since they give rise to convex constraints.  ...  GPU-based computation times of around 1 second allow for interactive segmentation.  ...  Shape Priors for Image Segmentation There has been much research on imposing prior shape knowledge into image segmentation.  ... 
doi:10.1109/iccv.2011.6126502 dblp:conf/iccv/KlodtC11 fatcat:5s5baj57ovchzjmxkmupqj34za

Convex multi-region probabilistic segmentation with shape prior in the isometric log-ratio transformation space

Shawn Andrews, Chris McIntosh, Ghassan Hamarneh
2011 2011 International Conference on Computer Vision  
Specifically, our energy function is convex and incorporates shape prior information while simultaneously generating a probabilistic segmentation for multiple regions.  ...  To our knowledge, these four goals (convex, with shape priors, multi-region, and probabilistic) do not exist together in any other method, and this is the first time ILR is used in an image segmentation  ...  Convex binary segmentation with shape prior Following [8] , we will define a convex energy functional as the sum of convex terms : E(q) = E Edge (q) + E Region (q) + E Shape (q) . (1) A binary segmentation  ... 
doi:10.1109/iccv.2011.6126484 dblp:conf/iccv/AndrewsMH11 fatcat:qhvk457xpvgfpk5fn2mmxtddsm

Fast Finsler Active Contours and Shape Prior Descriptor [chapter]

Foued Derraz, Abdelmalik Taleb-Ahmed, Laurent Peyrodie, Gerard Forzy, Christina Boydev
2011 Lecture Notes in Computer Science  
In this paper we proposed a new segmentation method based Fast Finsler Active Contours (FFAC).  ...  The FFAC is formulated in the Total Variation (TV) framework incorporating both region and shape descriptors.  ...  Conclusion We developed a fast globally segmentation based convex Finsler active contours model for binary segmentation in TV framework incorporating statistical and shape prior knowledge.  ... 
doi:10.1007/978-3-642-25085-9_22 fatcat:c7e4y4inkvextak7bikvprzdlu

A Probabilistic Interpretation Of Geometric Active Contour Segmentation

Jan Aelterman, Jonas De Vylder, Wilfried Philips, Dirk Van Haerenborgh
2014 Zenodo  
Winnok De Vos (Department of Molecular Biotechnology, Faculty of Bioscience Engineering, Ghent University) for sharing the fluorescent microscopic images and for manually generating ground truth.  ...  Segmentation prior The quality of the MAP estimator will depend on the prior used. While many shape priors exist for specific applications, we will restrict our priors to simple but generic priors.  ...  We used a Gaussian model for intensity modeling and simple shape priors such as area and perimeter. However other shape priors and other probability distributions could be used as well.  ... 
doi:10.5281/zenodo.43939 fatcat:6ulnnj5fvvdmdhjti25hsl4l3e

Shape priors in variational image segmentation: Convexity, Lipschitz continuity and globally optimal solutions

Daniel Cremers, Frank R. Schmidt, Frank Barthel
2008 2008 IEEE Conference on Computer Vision and Pattern Recognition  
Secondly, we prove that the introduction of shape priors into variational image segmentation leads to functionals which are convex with respect to shape deformations.  ...  For a large class of commonly considered (spatially continuous) functionals, we prove that -under mild regularity assumptions -segmentation and tracking with statistical shape priors can be performed in  ...  arbitrary spatial dimension (including shape priors for volumetric segmentation), and to a larger class of cost functionals and statistical shape priors.  ... 
doi:10.1109/cvpr.2008.4587446 dblp:conf/cvpr/CremersSB08 fatcat:dvjqc4z2bbeilg34brztbr6bve

Noise-Robust Pupil Center Detection through CNN-Based Segmentation with Shape-Prior Loss

Sang Yoon Han, Hyuk Jin Kwon, Yoonsik Kim, Nam Ik Cho
2020 IEEE Access  
INDEX TERMS Convex shape prior, deep learning, pupil segmentation, U-Net.  ...  In designing the loss function for the segmentation, we propose a new loss term that encodes the convex shape-prior for enhancing the robustness to noise.  ...  that encodes the convex shape prior.  ... 
doi:10.1109/access.2020.2985095 fatcat:fp75arzhy5bnzfzsl7qrrmqov4

A-expansion for multiple "hedgehog" shapes [article]

Hossam Isack, Yuri Boykov, Olga Veksler
2016 arXiv   pre-print
For example, star-convexity is popular for interactive single object segmentation due to simplicity and amenability to exact graph cut optimization.  ...  A single click and +/-90 degrees normal orientation constraints reduce our hedgehog prior to star-convexity. If all hedgehogs come from single clicks then our approach defines multi-star prior.  ...  In this case, we interpolate the vector field's orientation for every neighboring pixels and eliminate their edge constraint(s) if it were not consistent with the interpolated orientation, as shown in  ... 
arXiv:1602.01006v1 fatcat:aqsgrteznzauxlqecdmqf4llga

Star Convex Cuts with Encoding Swaps for Fast Whole-Spine Vertebrae Segmentation in MRI [article]

Marko Rak, Klaus D. Tönnies
2017 International Symposium on Vision, Modeling, and Visualization  
Our formulation involves appearance and shape information as well as star-convexity constraints to ensure a topologically correct segmentation for each vertebra.  ...  For close targets such as adjacent vertebrae, implementing star-convexity without fusing targets (naive binary formulations) or increasing run time/loosing optimality guarantees (multi-label formulations  ...  We contribute a novel binary graph cut formulation, which fuses patch-based star convex segmentation with whole-image constraints, ensuring topologically correctness for each and between vertebrae.  ... 
doi:10.2312/vmv.20171270 dblp:conf/vmv/RakT17 fatcat:kbxzvpjsurggzj6qfnyvaecqza
« Previous Showing results 1 — 15 out of 14,677 results