FROG

FROG Recognizer of Gestures

GestureHMM Class Reference

GestureHMM is a modeling class that represents a gesture. More...

Inheritance diagram for GestureHMM:
Inheritance graph
[legend]
Collaboration diagram for GestureHMM:
Collaboration graph
[legend]

List of all members.

Public Member Functions

 GestureHMM (String n)
 Constructs a GestureHMM with a name and default training parameters.
 GestureHMM (String n, Vector< Vector< Accel3D >> instances)
 Constructs a GestureHMM with a name and gesture set as well as default training parameters.
 GestureHMM (String n, GestureHMMParameters gParams)
 Constructs a GestureHMM with a name and training parameters.
 GestureHMM (String n, GestureHMMParameters gParams, Vector< Vector< Accel3D >> instances)
 Constructs a GestureHMM with a name, training parameters, as well as a gesture set.
 GestureHMM (String n, GestureHMMParameters gParams, KmeansParameters kParams)
 Constructs a GestureHMM with a name, training parameters, and quantizer parameters (Kmeans in particular).
 GestureHMM (String n, GestureHMMParameters gParams, KmeansParameters kParams, Vector< Vector< Accel3D >> instances)
 Constructs a GestureHMM with a name, training parameters, quantizer parameters (Kmeans in particular), and a training set.
 GestureHMM (String n, byte[] iconBuf) throws IOException
 Constructs a GestureHMM with a name and image.
 GestureHMM (String n, GestureHMMParameters gParams, byte[] iconBuf) throws IOException
 Constructs a GestureHMM with a name, training parameters, and an image.
 GestureHMM (String n, GestureHMMParameters gParams, Vector< Vector< Accel3D >> instances, byte[] iconBuf) throws IOException
 Constructs a GestureHMM with a name, training parameters, a gesture set, and an image.
 GestureHMM (String n, GestureHMMParameters gParams, KmeansParameters kParams, byte[] iconBuf) throws IOException
 Constructs a GestureHMM with a name, training parameters, a gesture set, and an image.
 GestureHMM (String n, GestureHMMParameters gParams, KmeansParameters kParams, Vector< Vector< Accel3D >> instances, byte[] iconBuf) throws IOException
 Constructs a GestureHMM with a name, training parameters, a gesture set, and an image.
boolean addAccel (int inum, Accel3D acc)
 Adds an additional Accel3D to the end of a gesture instance.
void addInstance (Vector< Accel3D > instance)
 Adds an entire gesture instance to the gesture set.
void clearStats ()
 Clears all saved statistics associated with this GestureModel.
void correct ()
 Signifies a correct recognition event to the GestureModel.
double getDefaultProbability ()
 For GestureHMM, default probability is obtained by averaging the probabilities with which each instance in the set used to create this model match the model.
BufferedImage getIcon ()
 Returns the icon of this gesture as a BufferedImage.
String getName ()
 Returns the name associated with this GestureModel.
int getNumCorrect ()
 Accessor for the number correct statistic.
int getNumIncorrect ()
 Accessor for the number incorrect statistic.
int getNumNotRecognized ()
 Accessor for the number not recognized statistic.
int getNumStates ()
 Returns the number of hidden states in this GestureHMM.
double getProbability (Vector< Accel3D > instance)
 For GestureHMM, probability is computed using the forward algorithm.
Vector< Vector< Accel3D > > getTrainingSet ()
 Accessor for the training set of gesture instances associated with this GestureModel.
double getTrainingThreshold ()
 Returns the training threshold of this GestureHMM.
void incorrect ()
 Signifies an incorrect recognition event to the GestureModel.
void notRecognized ()
 Signifies an unrecognized recognition event to the GestureModel.
boolean removeAccel (int inum, int anum)
 Removes an acceleration value from the gesture set.
boolean removeInstance (int inum)
 Removes a gesture instance from the gesture set.
void setName (String n)
 Associates a representative name with this GestureModel.
boolean train ()
 Trains the GestureModel, constructing everything necessary for recognition to take place.
boolean train (GestureHMMParameters gParams)
 Trains the GestureHMM with a specified set of training parameters.
boolean train (GestureHMMParameters gParams, KmeansParameters kParams)
 Trains the GestureHMM with a specified set of training parameters as well as a specified set of quantizer parameters.
boolean train (KmeansParameters params)
 Trains the GestureHMM with a specified set of quantizer parameters.
void setIcon (byte[] imageBuffer) throws IOException
 Associates a byte array image with the GestureModel.
double getAverageRecognitionProbability ()
 Returns the average certainty currently associated with this GestureModel.
void matchedWithProbability (double prob)
 Signifies a recognition event that occurred with a given probability.
int getType ()
 Accessor for the GestureModel type index.

Static Public Attributes

static final int GESTUREHMM = 0
 This is the GestureModel type index for GestureHMM models.
static final int NUM_TYPES_SUPPORTED = 1
 This is the number of GestureModel implementations currently supported by GestureModel.

Detailed Description

GestureHMM is a modeling class that represents a gesture.

As a model, it requires data (gesture instances) as well as parameters for training. GestureHMM utilizes a statistical HMM (as its name suggests) model as well as a Baum-Welch training method.

Author:
Team Better Recognize

Constructor & Destructor Documentation

GestureHMM ( String  n  ) 

Constructs a GestureHMM with a name and default training parameters.

Parameters:
n a name to associate with the GestureHMM
GestureHMM ( String  n,
Vector< Vector< Accel3D >>  instances 
)

Constructs a GestureHMM with a name and gesture set as well as default training parameters.

Parameters:
n a name to associate with the GestureHMM
instances the gesture set for training the GestureHMM
GestureHMM ( String  n,
GestureHMMParameters  gParams 
)

Constructs a GestureHMM with a name and training parameters.

Parameters:
n a name to associate with the GestureHMM
gParams HMM parameters for training

Here is the call graph for this function:

GestureHMM ( String  n,
GestureHMMParameters  gParams,
Vector< Vector< Accel3D >>  instances 
)

Constructs a GestureHMM with a name, training parameters, as well as a gesture set.

Parameters:
n a name to associate with the GestureHMM
gParams HMM parameters for training
instances the gesture set for training the GestureHMM
GestureHMM ( String  n,
GestureHMMParameters  gParams,
KmeansParameters  kParams 
)

Constructs a GestureHMM with a name, training parameters, and quantizer parameters (Kmeans in particular).

Parameters:
n a name to associate with the GestureHMM
gParams HMM parameters for training
kParams Kmeans parameters for training the quantizer

Here is the call graph for this function:

GestureHMM ( String  n,
GestureHMMParameters  gParams,
KmeansParameters  kParams,
Vector< Vector< Accel3D >>  instances 
)

Constructs a GestureHMM with a name, training parameters, quantizer parameters (Kmeans in particular), and a training set.

Parameters:
n a name to associate with the GestureHMM
gParams HMM parameters for training
kParams Kmeans parameters for training the quantizer
instances the gesture set for training the GestureHMM
GestureHMM ( String  n,
byte[]  iconBuf 
) throws IOException

Constructs a GestureHMM with a name and image.

Parameters:
n a name to associate with the GestureHMM
iconBuf the byte array image to associate with the GestureHMM
Exceptions:
IOException thrown if image reading fails
GestureHMM ( String  n,
GestureHMMParameters  gParams,
byte[]  iconBuf 
) throws IOException

Constructs a GestureHMM with a name, training parameters, and an image.

Parameters:
n a name to associate with the GestureHMM
gParams HMM parameters for training
iconBuf the byte array image to associate with the GestureHMM
Exceptions:
IOException thrown if image reading fails
GestureHMM ( String  n,
GestureHMMParameters  gParams,
Vector< Vector< Accel3D >>  instances,
byte[]  iconBuf 
) throws IOException

Constructs a GestureHMM with a name, training parameters, a gesture set, and an image.

Parameters:
n a name to associate with the GestureHMM
gParams HMM parameters for training
instances the gesture set for training the GestureHMM
iconBuf the byte array image to associate with the GestureHMM
Exceptions:
IOException thrown if image reading fails
GestureHMM ( String  n,
GestureHMMParameters  gParams,
KmeansParameters  kParams,
byte[]  iconBuf 
) throws IOException

Constructs a GestureHMM with a name, training parameters, a gesture set, and an image.

Parameters:
n a name to associate with the GestureHMM
gParams HMM parameters for training
kParams Kmeans parameters for training the quantizer
iconBuf the byte array image to associate with the GestureHMM
Exceptions:
IOException thrown if image reading fails
GestureHMM ( String  n,
GestureHMMParameters  gParams,
KmeansParameters  kParams,
Vector< Vector< Accel3D >>  instances,
byte[]  iconBuf 
) throws IOException

Constructs a GestureHMM with a name, training parameters, a gesture set, and an image.

Parameters:
n a name to associate with the GestureHMM
gParams HMM parameters for training
kParams Kmeans parameters for training the quantizer
instances the gesture set for training the GestureHMM
iconBuf the byte array image to associate with the GestureHMM
Exceptions:
IOException thrown if image reading fails

Member Function Documentation

boolean addAccel ( int  inum,
Accel3D  acc 
)

Adds an additional Accel3D to the end of a gesture instance.

Parameters:
inum The index of the gesture instance to modify
acc The Accel3D to insert
Returns:
a boolean signifying success or failure of the insertion

Implements GestureModel.

void addInstance ( Vector< Accel3D >  instance  ) 

Adds an entire gesture instance to the gesture set.

Parameters:
instance The instance to insert

Implements GestureModel.

void clearStats (  ) 

Clears all saved statistics associated with this GestureModel.

Implements GestureModel.

void correct (  ) 

Signifies a correct recognition event to the GestureModel.

Statistics for the model are updated accordingly.

Implements GestureModel.

double getAverageRecognitionProbability (  ) 

Returns the average certainty currently associated with this GestureModel.

The average certainty is calculated based on all calls made to frog.GestureModel#matchedWithProbability(double).

Returns:
the average recognition probability (certainty)

Implements GestureModel.

double getDefaultProbability (  ) 

For GestureHMM, default probability is obtained by averaging the probabilities with which each instance in the set used to create this model match the model.

In other words, it average the value of getProbability obtained with each gesture instance in the training set.

Implements GestureModel.

BufferedImage getIcon (  ) 

Returns the icon of this gesture as a BufferedImage.

Returns:
the icon of this gesture; or null if the gesture has no icon.

Implements GestureModel.

String getName (  ) 

Returns the name associated with this GestureModel.

(e.g. a model for a circle might appropriately be named "Circle")

Returns:
the name of the GestureModel.

Implements GestureModel.

int getNumCorrect (  ) 

Accessor for the number correct statistic.

This value is entirely dependent on calls to frog.GestureModel#correct()

Returns:
the saved number of correct recognition events.

Implements GestureModel.

int getNumIncorrect (  ) 

Accessor for the number incorrect statistic.

This value is entirely dependent on calls to frog.GestureModel#incorrect()

Returns:
the saved number of incorrect recognition events.

Implements GestureModel.

int getNumNotRecognized (  ) 

Accessor for the number not recognized statistic.

This value is entirely dependent on calls to frog.GestureModel#notRecognized()

Returns:
the saved number of unrecognized recognition events.

Implements GestureModel.

int getNumStates (  ) 

Returns the number of hidden states in this GestureHMM.

Returns:
the number of hidden states in this GestureHMM.

Here is the caller graph for this function:

double getProbability ( Vector< Accel3D >  instance  ) 

For GestureHMM, probability is computed using the forward algorithm.

Implements GestureModel.

Here is the call graph for this function:

Vector<Vector<Accel3D> > getTrainingSet (  ) 

Accessor for the training set of gesture instances associated with this GestureModel.

Returns:
the model's gesture set

Implements GestureModel.

double getTrainingThreshold (  ) 

Returns the training threshold of this GestureHMM.

Returns:
the training threshold of this GestureHMM.

Here is the caller graph for this function:

int getType (  ) 

Accessor for the GestureModel type index.

Should return the value as specified in the GestureModel interface.

Returns:
the GestureModel type index.

Implements GestureModel.

void incorrect (  ) 

Signifies an incorrect recognition event to the GestureModel.

Statistics for the model are updated accordingly.

Implements GestureModel.

void matchedWithProbability ( double  prob  ) 

Signifies a recognition event that occurred with a given probability.

This updates the average recognition probability of the GestureModel, as accessed by frog.GestureModel#getAverageRecognitionProbability()

Parameters:
prob the probability (0-1) with which the match occurred

Implements GestureModel.

void notRecognized (  ) 

Signifies an unrecognized recognition event to the GestureModel.

Statistics for the model are updated accordingly.

Implements GestureModel.

boolean removeAccel ( int  inum,
int  anum 
)

Removes an acceleration value from the gesture set.

Parameters:
inum the index of the instance to access
anum the index of the acceleration to remove
Returns:
a boolean signifying success or failure of the deletion

Implements GestureModel.

boolean removeInstance ( int  inum  ) 

Removes a gesture instance from the gesture set.

Parameters:
inum the index of the instance to remove
Returns:
a boolean signifying success or failure of the deletion

Implements GestureModel.

void setIcon ( byte[]  imageBuffer  )  throws IOException

Associates a byte array image with the GestureModel.

Parameters:
imageBuffer the byte array image
Exceptions:
IOException thrown if image reading fails

Implements GestureModel.

void setName ( String  n  ) 

Associates a representative name with this GestureModel.

Parameters:
n the name to associate

Implements GestureModel.

boolean train ( KmeansParameters  params  ) 

Trains the GestureHMM with a specified set of quantizer parameters.

The Quantizer for the GestureHMM is set accordingly.

Parameters:
params The K-means parameters.
Returns:
A boolean indicating whether training took place or not. Training will not occur if the gesture set is empty.
boolean train ( GestureHMMParameters  gParams,
KmeansParameters  kParams 
)

Trains the GestureHMM with a specified set of training parameters as well as a specified set of quantizer parameters.

The Quantizer for the GestureHMM is set accordingly and the training parameters associated with the GestureHMM are set to the new values in gParams.

Parameters:
gParams the training parameters
kParams the Kmeans parameters
Returns:
a boolean indicating whether training took place or not. Training will not occur if the gesture set is empty.

Here is the call graph for this function:

boolean train ( GestureHMMParameters  gParams  ) 

Trains the GestureHMM with a specified set of training parameters.

The training parameters associated with the GestureHMM are set to the new values in gParams.

Parameters:
gParams the training parameters
Returns:
a boolean indicating whether training took place or not. Training will not occur if the gesture set is empty.

Here is the call graph for this function:

boolean train (  ) 

Trains the GestureModel, constructing everything necessary for recognition to take place.

Implements GestureModel.

Here is the caller graph for this function:

Member Data Documentation

final int GESTUREHMM = 0 [static, inherited]

This is the GestureModel type index for GestureHMM models.

final int NUM_TYPES_SUPPORTED = 1 [static, inherited]

This is the number of GestureModel implementations currently supported by GestureModel.

The documentation for this class was generated from the following file:
  • /Users/dev/Documents/SVN brazos.cs.tcu.edu/trunk/FROG/src/frog/GestureHMM.java
 All Classes Namespaces Files Functions Variables