/*
* This file is part of the LIRE project: http://lire-project.net
* LIRE is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* LIRE is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with LIRE; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
* We kindly ask you to refer the any or one of the following publications in
* any publication mentioning or employing Lire:
*
* Lux Mathias, Savvas A. Chatzichristofis. Lire: Lucene Image Retrieval -
* An Extensible Java CBIR Library. In proceedings of the 16th ACM International
* Conference on Multimedia, pp. 1085-1088, Vancouver, Canada, 2008
* URL: http://doi.acm.org/10.1145/1459359.1459577
*
* Lux Mathias. Content Based Image Retrieval with LIRE. In proceedings of the
* 19th ACM International Conference on Multimedia, pp. 735-738, Scottsdale,
* Arizona, USA, 2011
* URL: http://dl.acm.org/citation.cfm?id=2072432
*
* Mathias Lux, Oge Marques. Visual Information Retrieval using Java and LIRE
* Morgan & Claypool, 2013
* URL: http://www.morganclaypool.com/doi/abs/10.2200/S00468ED1V01Y201301ICR025
*/
package net.semanticmetadata.lire.aggregators;
import net.semanticmetadata.lire.classifiers.Cluster;
import net.semanticmetadata.lire.imageanalysis.features.LocalFeature;
import net.semanticmetadata.lire.utils.SerializationUtils;
import java.util.Arrays;
import java.util.List;
/**
* General class creating vector representations based on VLAD model, given the list of local features and the codebook.
* Created by Nektarios on 03/06/2015.
*
* @author Nektarios Anagnostopoulos, nek.anag@gmail.com
* (c) 2015 by Nektarios Anagnostopoulos
*/
public class VLAD extends AbstractAggregator {
private double[] vector;
public VLAD() { }
/**
* Given a list of features and a codebook, {@link VLAD#createVectorRepresentation(List, Cluster[])} aggregates
* the features to create the vector representation according to the VLAD model.
* @param listOfLocalFeatures is the list of features.
* @param clustersArray is the codebook.
*/
@Override
public void createVectorRepresentation(List<? extends LocalFeature> listOfLocalFeatures, Cluster[] clustersArray) {
vector = new double[clustersArray.length * (clustersArray[0].getMean()).length];
Arrays.fill(vector, 0d);
int clusterIndex;
double[] mean;
// VLAD - Vector of Locally Aggregated Descriptors
for (LocalFeature localFeature : listOfLocalFeatures) {
clusterIndex = clusterForFeature(localFeature.getFeatureVector(), clustersArray);
mean = clustersArray[clusterIndex].getMean();
for (int i = 0; i < localFeature.getFeatureVector().length; i++) {
vector[clusterIndex * localFeature.getFeatureVector().length + i] += (localFeature.getFeatureVector()[i] - mean[i]);
}
}
normalize(vector);
}
/**
* Returns the vector representation in byte[] format.
* @return the vector representation as a byte array.
*/
@Override
public byte[] getByteVectorRepresentation() { return SerializationUtils.toByteArray(vector); }
/**
* Returns the vector representation in string format.
* @return the vector representation as string.
*/
@Override
public String getStringVectorRepresentation() { return SerializationUtils.toString(vector); }
/**
* Returns the vector representation in double[] format.
* @return the vector representation as a double array.
*/
@Override
public double[] getVectorRepresentation() { return vector; }
@Override
public String getFieldName() {
return Aggregator.FIELD_NAME_VLAD;
}
/**
* Returns the vector representation in string format.
* @return the vector representation as string.
*/
public String toString() { return getStringVectorRepresentation();}
private void normalize(double[] histogram) {
// L2
double sumOfSquares = 0;
for (double next : histogram) {
sumOfSquares += (next * next);
}
if (sumOfSquares > 0) {
sumOfSquares = Math.sqrt(sumOfSquares);
for (int i = 0; i < histogram.length; i++) {
// histogram[i] = Math.floor(16d * histogram[i] / sumOfSquares);
histogram[i] /= sumOfSquares;
}
}
/* // L1
double min = Double.MAX_VALUE, max = Double.MIN_VALUE;
for (int i = 0; i < histogram.length; i++) {
min = Math.min(histogram[i], min);
max = Math.max(histogram[i], max);
}
for (int i = 0; i < histogram.length; i++) {
histogram[i] = (histogram[i] - min) / (max - min);
}*/
}
}