org.openimaj.ml.annotation.linear

Class DenseLinearTransformAnnotator<OBJECT,ANNOTATION>

java.lang.Object

org.openimaj.ml.annotation.AbstractAnnotator<OBJECT,ANNOTATION>

org.openimaj.ml.annotation.BatchAnnotator<OBJECT,ANNOTATION>

org.openimaj.ml.annotation.linear.DenseLinearTransformAnnotator<OBJECT,ANNOTATION>

Type Parameters:

OBJECT - Type of object being annotated

ANNOTATION - Type of annotation

All Implemented Interfaces:

Classifier<ANNOTATION,OBJECT>, Annotator<OBJECT,ANNOTATION>, BatchTrainer<Annotated<OBJECT,ANNOTATION>>

@Reference(type=Inproceedings,
           author={"Jonathan Hare","Paul Lewis"},
           title="Semantic Retrieval and Automatic Annotation: Linear Transformations, Correlation and Semantic Spaces",
           year="2010",
           booktitle="Imaging and Printing in a Web 2.0 World; and Multimedia Content Access: Algorithms and Systems IV",
           url="http://eprints.soton.ac.uk/268496/",
           note=" Event Dates: 17-21 Jan 2010",
           month="January",
           publisher="SPIE",
           volume="7540")
public class DenseLinearTransformAnnotator<OBJECT,ANNOTATION>
extends BatchAnnotator<OBJECT,ANNOTATION>

An annotator that determines a "transform" between feature vectors and vectors of annotation counts. The transform is estimated using a lossy pseudo inverse; the single parameter of the algorithm is the desired rank of the transform matrix.

Author:

Jonathon Hare (jsh2@ecs.soton.ac.uk)

Field Summary

Fields

Modifier and Type	Field and Description
protected int	k
protected List<ANNOTATION>	terms
protected Jama.Matrix	transform

Constructor Summary

Constructors

Constructor and Description
DenseLinearTransformAnnotator(int k, FeatureExtractor<? extends FeatureVector,OBJECT> extractor) Construct with the given number of dimensions and feature extractor.

Method Summary

Methods

Modifier and Type	Method and Description
List<ScoredAnnotation<ANNOTATION>>	annotate(OBJECT image) Generate annotations for the given object.
Set<ANNOTATION>	getAnnotations()
void	train(List<? extends Annotated<OBJECT,ANNOTATION>> data) Train the object with the given data.

Methods inherited from class org.openimaj.ml.annotation.BatchAnnotator

train

Methods inherited from class org.openimaj.ml.annotation.AbstractAnnotator

classify

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

terms

protected List<ANNOTATION> terms

transform

protected Jama.Matrix transform

k

protected int k

Constructor Detail

DenseLinearTransformAnnotator

public DenseLinearTransformAnnotator(int k,
                             FeatureExtractor<? extends FeatureVector,OBJECT> extractor)

Construct with the given number of dimensions and feature extractor.

Parameters:

k - the number of dimensions (rank of the pseudo-inverse)

extractor - the feature extractor

Method Detail

train

public void train(List<? extends Annotated<OBJECT,ANNOTATION>> data)

Description copied from interface: BatchTrainer

Train the object with the given data. BatchTrainers should reset the underlying model everytime this method is called.

Parameters:

data - the training data

annotate

public List<ScoredAnnotation<ANNOTATION>> annotate(OBJECT image)

Description copied from interface: Annotator

Generate annotations for the given object.

Parameters:

image - the image

Returns:

generated annotations

getAnnotations

public Set<ANNOTATION> getAnnotations()

Returns:

a Set of all annotations this Annotator knows about