Cover Image

High-order models in semantic image segmentation /

High-Order Models in Semantic Image Segmentation reviews recent developments in optimization-based methods for image segmentation, presenting several geometric and mathematical models that underlie a broad class of recent segmentation techniques. Focusing on impactful algorithms in the computer visi...

Full description

Bibliographic Details
Main Author: Ben Ayed, Ismail (Author)
Corporate Author: ScienceDirect (Online service)
Format: eBook
Language:English
Published: London : Academic Press, [2023]
Subjects:
Image segmentation.
Image segmentation > Mathematical models.
Semantic computing.
Segmentation d'image.
Segmentation d'image > Modèles mathématiques.
Informatique sémantique.
Electronic books.
Online Access:Connect to the full text of this electronic book
Table of Contents:
  • Intro
  • Title page
  • Table of Contents
  • Copyright
  • General introduction
  • General context
  • From graphical models to deep learning
  • Chapter 1
  • Chapter 2
  • Chapter 3
  • Chapter 4
  • Chapter 5
  • Chapter 6
  • Chapter 7
  • Chapter 8
  • Chapter 9
  • Chapter 10
  • Chapter 1: Markov random fields
  • Abstract
  • 1.1. Discrete representations
  • 1.2. Popular optimizers for random fields
  • References
  • Chapter 2: Graph cuts
  • Abstract
  • 2.1. Min-cut and max-flow problems
  • 2.2. Move-making algorithms for multi-label problems
  • References
  • Chapter 3: Mean-field inference
  • Abstract
  • 3.1. Pairwise conditional random field functions
  • 3.2. Mean-field inference
  • Appendix 3.A.
  • References
  • Chapter 4: Regularized model fitting
  • Abstract
  • 4.1. General probabilistic form
  • 4.2. Standard models
  • References
  • Chapter 5: Regularized mutual information
  • Abstract
  • 5.1. Model fitting as entropy minimization
  • 5.2. Limitations of entropy and highly descriptive models
  • 5.3. A discriminative view of the mutual information
  • References
  • Chapter 6: Examples of high-order functionals
  • Abstract
  • 6.1. Introduction
  • 6.2. Shape priors
  • 6.3. Graph clustering
  • 6.4. Distribution matching
  • References
  • Chapter 7: Pseudo-bound optimization
  • Abstract
  • 7.1. Bound optimization
  • 7.2. Bound optimization
  • 7.3. Pseudo-bound optimization
  • 7.4. Auxiliary functionals
  • References
  • Chapter 8: Trust-region optimization
  • Abstract
  • 8.1. General-form problem
  • 8.2. Trust-region optimization
  • 8.3. A shape prior example
  • 8.4. Details of the Gateâux derivatives
  • References
  • Chapter 9: Random field losses for deep networks
  • Abstract
  • 9.1. Fully supervised segmentation
  • 9.2. Weakly supervised segmentation
  • 9.3. Beyond gradient descent for random field losses
  • References.
  • Chapter 10: Constrained deep networks
  • Abstract
  • 10.1. Weakly supervised segmentation via constrained CNNs
  • 10.2. Constraint optimization
  • 10.3. Discussion of some experimental results
  • References
  • Index.