Corpus ID: 13907471

Closed-Form Training of Conditional Random Fields for Large Scale Image Segmentation

@article{Kolesnikov2014ClosedFormTO,
  title={Closed-Form Training of Conditional Random Fields for Large Scale Image Segmentation},
  author={Alexander Kolesnikov and M. Guillaumin and V. Ferrari and Christoph H. Lampert},
  journal={ArXiv},
  year={2014},
  volume={abs/1403.7057}
}
  • Alexander Kolesnikov, M. Guillaumin, +1 author Christoph H. Lampert
  • Published 2014
  • Computer Science
  • ArXiv
  • We present LS-CRF, a new method for very efficient large-scale training of Conditional Random Fields (CRFs). It is inspired by existing closed-form expressions for the maximum likelihood parameters of a generative graphical model with tree topology. LS-CRF training requires only solving a set of independent regression problems, for which closed-form expression as well as efficient iterative solvers are available. This makes it orders of magnitude faster than conventional maximum likelihood… CONTINUE READING
    10 Citations
    Marginal Weighted Maximum Log-likelihood for Efficient Learning of Perturb-and-Map models
    • 2
    • Highly Influenced
    • PDF
    Efficient Piecewise Training of Deep Structured Models for Semantic Segmentation
    • 661
    • PDF
    Accurate 3D bone segmentation in challenging CT images: Bottom-up parsing and contextualized optimization
    • 3
    • PDF
    Exploring Context with Deep Structured Models for Semantic Segmentation
    • 90
    • PDF
    Conditional Random Field and Deep Feature Learning for Hyperspectral Image Classification
    • 21
    • PDF
    CNN-based Segmentation of Medical Imaging Data
    • 160
    • PDF

    References

    SHOWING 1-10 OF 31 REFERENCES
    Decision tree fields
    • 121
    • PDF
    Efficient training for pairwise or higher order CRFs via dual decomposition
    • 54
    • PDF
    Piecewise Training for Undirected Models
    • 194
    • PDF
    Structured Learning and Prediction in Computer Vision
    • 307
    • PDF
    Segmentation Propagation in ImageNet
    • 617
    • PDF
    TextonBoost for Image Understanding: Multi-Class Object Recognition and Segmentation by Jointly Modeling Texture, Layout, and Context
    • 1,038
    • PDF
    Object Class Segmentation using Random Forests
    • 229
    • PDF
    Decomposing a scene into geometric and semantically consistent regions
    • 669
    • PDF
    Graphical Models, Exponential Families, and Variational Inference
    • 3,306
    • PDF