org.openimaj.image.processing.face.detection.keypoints

Class FacialKeypointExtractor

java.lang.Object

org.openimaj.image.processing.face.detection.keypoints.FacialKeypointExtractor

@Reference(type=Inproceedings,
           author={"Mark Everingham","Josef Sivic","Andrew Zisserman"},
           title="Hello! My name is... Buffy - Automatic naming of characters in TV video",
           year="2006",
           booktitle="In BMVC")
public class FacialKeypointExtractor
extends Object

A class capable of finding likely facial keypoints using a masked Haar cascade for each keypoint, and then picking the best combination of points based on a model.

Implementation and data is based on Mark Everingham's Oxford VGG Baseline Face Processing Code

Author:

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

Field Summary

Fields

Modifier and Type	Field and Description
protected org.openimaj.image.processing.face.detection.keypoints.FacialKeypointExtractor.Model	model

Constructor Summary

Constructors

Constructor and Description
FacialKeypointExtractor() Construct a new facial keypoint extractor.

Method Summary

Modifier and Type	Method and Description
FacialKeypoint[]	extractFacialKeypoints(FImage canonicalImage) Extract the facial keypoints from a canonical image.
protected FacialKeypoint[]	findParts(FImage[] AC)
int	getCanonicalImageDimension() Get the size of the face image that the keypoint extractor can work with.
int	hashCode()

Methods inherited from class java.lang.Object

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

Field Detail

model

protected org.openimaj.image.processing.face.detection.keypoints.FacialKeypointExtractor.Model model

Constructor Detail

FacialKeypointExtractor

public FacialKeypointExtractor()

Construct a new facial keypoint extractor.

Method Detail

getCanonicalImageDimension

public int getCanonicalImageDimension()

Get the size of the face image that the keypoint extractor can work with.

Returns:

the size (both height and width)

extractFacialKeypoints

public FacialKeypoint[] extractFacialKeypoints(FImage canonicalImage)

Extract the facial keypoints from a canonical image. The image must contain a face and have the dimensions specified by getCanonicalImageDimension().

Parameters:

canonicalImage - the canonical image containing a detected face.

Returns:

A list of facial keypoints for the image.

findParts

protected FacialKeypoint[] findParts(FImage[] AC)

hashCode

public int hashCode()

Overrides:

hashCode in class Object