Databases

Wrapper classes for image databases

class facereclib.databases.Database(name, original_directory=None, original_extension=None, annotation_directory=None, annotation_extension='.pos', annotation_type=None, protocol='Default', **kwargs)

This class represents the basic API for database access. Please use this class as a base class for your database access classes. Do not forget to call the constructor of this base class in your derived class.

Parameters to the constructor of the Database:

name

A unique name for the database.

original_directory

The directory where the original data of the database are stored.

original_extension

The file extension of the original data.

annotation_directory

The directory where the image annotations of the database are stored, if any.

annotation_directory

The file extension of the annotation files.

protocol

The name of the protocol that defines the default experimental setup for this database.

kwargs

Ignored extra arguments.

all_files(groups=None)

Returns all files of the database

annotations(file)

Returns the annotations for the given File object, if available.

arrange_by_client(files)

Arranges the given list of files by client id. This function returns a list of lists of File’s.

client_id_from_model_id(model_id, group='dev')

Returns the client id for the given model id

enroll_files(model_id, group='dev')

Returns a list of enrollment File objects for the given model id and the given group

file_names(files, directory, extension)

Returns the full path of the given File objects.

model_ids(group='dev')

Returns a list of model ids for the given group

original_file_names(files)

Returns the full path of the original data of the given File objects.

probe_file_sets(model_id=None, group='dev')

Returns a list of probe FileSet object in a specific format that should be compared with the model belonging to the given model id of the specified group

probe_files(model_id=None, group='dev')

Returns a list of probe File object in a specific format that should be compared with the model belonging to the given model id of the specified group

sort(files)

Returns a sorted version of the given list of File’s (or other structures that define an ‘id’ data member). The files will be sorted according to their id, and duplicate entries will be removed.

training_files(step=None, arrange_by_client=False)

Returns all training File objects for the given step (might be ‘train_extractor’, ‘train_projector’, ‘train_enroller’ or None), and arranges them by client, if desired

uses_probe_file_sets()

Defines if, for the current protocol, the database uses several probe files to generate a score. By default, False is returned. Overwrite the default if you need different behavior.

class facereclib.databases.DatabaseBob(database, all_files_options={}, extractor_training_options={}, projector_training_options={}, enroller_training_options={}, check_original_files_for_existence=False, **kwargs)

Bases: facereclib.databases.Database.Database

This class can be used whenever you have a database that follows the default Bob database interface.

Parameters of the constructor of this database:

database

the bob.db.___ database that provides the actual interface

image_directory

The directory where the original images are stored.

image_extension

The file extension of the original images.

all_files_options

Options passed to the database query used to retrieve all data.

extractor_training_options

Options passed to the database query used to retrieve the images for the extractor training.

projector_training_options

Options passed to the database query used to retrieve the images for the projector training.

enroller_training_options

Options passed to the database query used to retrieve the images for the enroller training.

check_original_files_for_existence

Enables the test for the original data files when querying the database.

kwargs

The arguments of the base class

all_files(groups=None)

Returns all File objects of the database for the current protocol. If the current protocol is ‘None’ (a string), None (NoneType) will be used instead

annotations(file)

Returns the annotations for the given File object, if available.

arrange_by_client(files)

Arranges the given list of files by client id. This function returns a list of lists of File’s.

client_id_from_model_id(model_id, group='dev')

Returns the client id for the given model id.

enroll_files(model_id, group='dev')

Returns the list of enrollment File objects for the given model id.

file_names(files, directory, extension)

Returns the full path of the given File objects.

model_ids(group='dev')

Returns the model ids for the given group and the current protocol.

original_file_names(files)

Returns the full path of the original data of the given File objects.

probe_file_sets(model_id=None, group='dev')

Returns the list of probe File objects (for the given model id, if given).

probe_files(model_id=None, group='dev')

Returns the list of probe File objects (for the given model id, if given).

sort(files)

Returns a sorted version of the given list of File’s (or other structures that define an ‘id’ data member). The files will be sorted according to their id, and duplicate entries will be removed.

test_files(groups=['dev'])

Returns the test files (i.e., enrollment and probe files) for the given groups.

training_files(step=None, arrange_by_client=False)

Returns all training File objects of the database for the current protocol.

uses_probe_file_sets()

Defines if, for the current protocol, the database uses several probe files to generate a score.

class facereclib.databases.DatabaseBobZT(z_probe_options={}, **kwargs)

Bases: facereclib.databases.DatabaseBob.DatabaseBob, facereclib.databases.Database.DatabaseZT

This class can be used whenever you have a database that follows the default Bob database interface defining file lists for ZT score normalization.

all_files(groups=['dev'])

Returns all File objects of the database for the current protocol. If the current protocol is ‘None’ (a string), None (NoneType) will be used instead

annotations(file)

Returns the annotations for the given File object, if available.

arrange_by_client(files)

Arranges the given list of files by client id. This function returns a list of lists of File’s.

client_id_from_model_id(model_id, group='dev')

Returns the client id for the given model id.

client_id_from_t_model_id(t_model_id, group='dev')

Returns the client id for the given T-model id. In this base class implementation, we just use the client_id_from_model_id() function. Overload this function if you need another behavior.

enroll_files(model_id, group='dev')

Returns the list of enrollment File objects for the given model id.

file_names(files, directory, extension)

Returns the full path of the given File objects.

model_ids(group='dev')

Returns the model ids for the given group and the current protocol.

original_file_names(files)

Returns the full path of the original data of the given File objects.

probe_file_sets(model_id=None, group='dev')

Returns the list of probe File objects (for the given model id, if given).

probe_files(model_id=None, group='dev')

Returns the list of probe File objects (for the given model id, if given).

sort(files)

Returns a sorted version of the given list of File’s (or other structures that define an ‘id’ data member). The files will be sorted according to their id, and duplicate entries will be removed.

t_enroll_files(model_id, group='dev')

Returns the list of enrollment File objects for the given T-Norm model id.

t_model_ids(group='dev')

Returns the T-Norm model ids for the given group and the current protocol.

test_files(groups=['dev'])

Returns the test files (i.e., enrollment and probe files) for the given groups.

training_files(step=None, arrange_by_client=False)

Returns all training File objects of the database for the current protocol.

uses_probe_file_sets()

Defines if, for the current protocol, the database uses several probe files to generate a score.

z_probe_file_sets(group='dev')

Returns the list of Z-probe Fileset objects.

z_probe_files(group='dev')

Returns the list of Z-probe File objects.

class facereclib.databases.DatabaseFileList(database, **kwargs)

Bases: facereclib.databases.DatabaseBob.DatabaseBobZT

This class should be used whenever you have an bob.db.verification.filelist.Database`.

Parameters of the constructor of this database:

database : bob.db.verification.filelist.Database

The database that provides the actual interface

kwargs

Keyword arguments directly passed to the DatabaseBobZT base class constructor

all_files(groups=['dev'])

Returns all File objects of the database for the current protocol. If the current protocol is ‘None’ (a string), None (NoneType) will be used instead

annotations(file)

Returns the annotations for the given File object, if available.

arrange_by_client(files)

Arranges the given list of files by client id. This function returns a list of lists of File’s.

client_id_from_model_id(model_id, group='dev')

Returns the client id for the given model id.

client_id_from_t_model_id(t_model_id, group='dev')

Returns the client id for the given T-model id.

enroll_files(model_id, group='dev')

Returns the list of enrollment File objects for the given model id.

file_names(files, directory, extension)

Returns the full path of the given File objects.

model_ids(group='dev')

Returns the model ids for the given group and the current protocol.

original_file_names(files)

Returns the full path of the original data of the given File objects.

probe_file_sets(model_id=None, group='dev')

Returns the list of probe File objects (for the given model id, if given).

probe_files(model_id=None, group='dev')

Returns the list of probe File objects (for the given model id, if given).

sort(files)

Returns a sorted version of the given list of File’s (or other structures that define an ‘id’ data member). The files will be sorted according to their id, and duplicate entries will be removed.

t_enroll_files(model_id, group='dev')

Returns the list of enrollment File objects for the given T-Norm model id.

t_model_ids(group='dev')

Returns the T-Norm model ids for the given group and the current protocol.

test_files(groups=['dev'])

Returns the test files (i.e., enrollment and probe files) for the given groups.

training_files(step=None, arrange_by_client=False)

Returns all training File objects of the database for the current protocol.

uses_probe_file_sets()

Defines if, for the current protocol, the database uses several probe files to generate a score.

z_probe_file_sets(group='dev')

Returns the list of Z-probe Fileset objects.

z_probe_files(group='dev')

Returns the list of Z-probe File objects.

class facereclib.databases.DatabaseZT(name, original_directory=None, original_extension=None, annotation_directory=None, annotation_extension='.pos', annotation_type=None, protocol='Default', **kwargs)

Bases: facereclib.databases.Database.Database

This class defines additional API functions that are required to compute ZT score normalization. During construction, please call the constructor of the base class ‘Database’ directly.

Parameters to the constructor of the Database:

name

A unique name for the database.

original_directory

The directory where the original data of the database are stored.

original_extension

The file extension of the original data.

annotation_directory

The directory where the image annotations of the database are stored, if any.

annotation_directory

The file extension of the annotation files.

protocol

The name of the protocol that defines the default experimental setup for this database.

kwargs

Ignored extra arguments.

all_files(groups=None)

Returns all files of the database

annotations(file)

Returns the annotations for the given File object, if available.

arrange_by_client(files)

Arranges the given list of files by client id. This function returns a list of lists of File’s.

client_id_from_model_id(model_id, group='dev')

Returns the client id for the given model id

client_id_from_t_model_id(t_model_id, group='dev')

Returns the client id for the given T-model id. In this base class implementation, we just use the client_id_from_model_id() function. Overload this function if you need another behavior.

enroll_files(model_id, group='dev')

Returns a list of enrollment File objects for the given model id and the given group

file_names(files, directory, extension)

Returns the full path of the given File objects.

model_ids(group='dev')

Returns a list of model ids for the given group

original_file_names(files)

Returns the full path of the original data of the given File objects.

probe_file_sets(model_id=None, group='dev')

Returns a list of probe FileSet object in a specific format that should be compared with the model belonging to the given model id of the specified group

probe_files(model_id=None, group='dev')

Returns a list of probe File object in a specific format that should be compared with the model belonging to the given model id of the specified group

sort(files)

Returns a sorted version of the given list of File’s (or other structures that define an ‘id’ data member). The files will be sorted according to their id, and duplicate entries will be removed.

t_enroll_files(model_id, group='dev')

Returns a list of enrollment files for the given T-Norm model id and the given group

t_model_ids(group='dev')

Returns a list of T-Norm model ids for the given group

training_files(step=None, arrange_by_client=False)

Returns all training File objects for the given step (might be ‘train_extractor’, ‘train_projector’, ‘train_enroller’ or None), and arranges them by client, if desired

uses_probe_file_sets()

Defines if, for the current protocol, the database uses several probe files to generate a score. By default, False is returned. Overwrite the default if you need different behavior.

z_probe_file_sets(model_id=None, group='dev')

Returns a list of Z-probe FileSets object in a specific format that should be compared with the model belonging to the given model id of the specified group

z_probe_files(model_id=None, group='dev')

Returns a list of Z-probe objects in a specific format that should be compared with the model belonging to the given model id of the specified group

class facereclib.databases.File(file_id, client_id, path)

This class defines the minimum interface of a file that needs to be exported

class facereclib.databases.FileSet(file_set_id, client_id, file_set_name)

This class defines the minimum interface of a file set that needs to be exported