Marginal Space Learning

Adrian Barbu

Learning a classifier in a marginal space in which some of the object parameters are integrated out (ignored) allows to elliminate thousands of uninteresting locations of the object parameter space in one step. This way speedups by 3-6 orders of magnitude have been observed.
This principle was sucessfully applied to many medical imaging projects including Guidewire Localization and Heart Segmentation.Hear 3D Mesh

  1. Y. Zheng, Adrian Barbu, B. Georgescu, M. Scheuering and D. Comaniciu. Four-Chamber Heart Modeling and
    Automatic Segmentation for 3D Cardiac CT Volumes Using Marginal Space Learning and Steerable Features.
    IEEE Trans Medical Imaging    , November 2008. (pdf)
  2. L. Lu, Adrian Barbu, J. Liang, L. Bogoni, M. Salganicoff and D. Comaniciu. Simultaneous Detection and Registration
    for Ileo-Cecal Valve Detection in 3D CT Colonography    . ECCV 2008 
  3. L. Lu, Adrian Barbu, M. Wolf, J. Liang, M. Salganicoff, D. Comaniciu. Accurate Polyp Segmentation for 3D 
    CT Colonography Using Multi-Staged Probabilistic Binary Learning and Compositional Model. CVPR 2008.(pdf)
  4. R. Socher, Adrian Barbu, D. Comaniciu. A Learning Based Hierarchical Model for Vessel Segmentation. 
    IEEE International Symposium on Biomedical Imaging, 2008. (pdf)
  5. Y. Zheng, Adrian Barbu, B. Georgescu, M. Scheuering, D. Comaniciu. Fast Automatic Heart Chamber Segmentation
    from 3D CT Data Using Marginal Space Learning and Steerable Features. ICCV 2007 (pdf)
  6. S Lakare, M Wolf, L Bogoni, Adrian Barbu, M Dundar, L Lu, M Salganicoff, Evaluation of a Learning-based
    Component for Suppression of False Positives Located on the Ileo Cecal Valve or Rectal Tube, RSNA 2007
  7. A. Barbu, V. Athitsos, B. Georgescu, S. Boehm, P. Durlak, D. Comaniciu. Hierarchical Learning of Curves:
    Application to Guidewire Localization in Fluoroscopy. CVPR 2007 (pdf)
home    research