Visual learning related methods and concepts

A part of the CVonline computer vision resource summarizing some of the different methods for learning models, model parameters, or algorithmic parameters, as commonly used in computer vision and image processing.

Observational learning
1. Discrete observational learning
2. Probabilistic observational learning
Geometric feature learning
Joint natural language and image data learning

1. NLP learning techniques

Learning technologies

1. Bayesian learning/Probabilistic model learning
  1. Bayesian principal component analysis
  2. Latent variable learning
  3. Variational Bayesian methods
2. Clustering

- 1. Clustering coefficient
  2. Clustering quality evaluation (See also Generic methods->Model Selection Criteria)
    1. Dunn Clustering Index
    2. Silhouette Clustering Index
  3. Fuzzy clustering
  4. Hierarchical clustering
  5. k-means clustering

- - 1. Hierarchical k-means clustering

1. Gaussian mixture models, Expectation-Maximization (EM)
2. Ensemble learning

- 1. Bootstrap aggregating
  2. Boosting
    1. AdaBoost
    2. DenseBoost
    3. TextonBoost
  3. Extremely random trees (Extra-trees)
  4. Random forests
  5. Vector boosting

1. Feature selection
2. Gaussian process learning and classification
3. Genetic programming/Genetic algorithms
4. Neural networks
5. Principal component analysis
6. Support vector machines

- 1. Exemplar SVM
  2. Kernel methods
  3. Kernel trick
  4. Structured SVM
  5. Relevance vector machine

1. Semi-supervised learning
2. Vector quantization

Shape model learning

- 1. Marching cubes

Property learning

1. Spatio-temporal patterns

Google Sites

Report abuse