BEAT - tpereira/gmm_statistics

This algorithm is a legacy one. The API has changed since its implementation. New versions and forks will need to be updated.

This algorithm is splittable

Endpoint Groups 1

Algorithms have at least one input and one output. All algorithm endpoints are organized in groups. Groups are used by the platform to indicate which inputs and outputs are synchronized together. The first group is automatically synchronized with the channel defined by the block in which the algorithm is deployed.

Unnamed group

Endpoint Name	Data Format	Nature
ubm	tutorial/gmm/1	Input
features	system/array_2d_floats/1	Input
statistics	tutorial/gmm_statistics/1	Output

import bob
import numpy
from bob.machine import GMMMachine

def gmm_from_data(data):
    """Unmangles a bob.machine.GMMMachine from a BEAT Data object"""

dim_c, dim_d = data.means.shape
    gmm = GMMMachine(dim_c, dim_d)
    gmm.weights = data.weights
    gmm.means = data.means
    gmm.variances = data.variances
    gmm.variance_thresholds = data.variance_thresholds
    return gmm

class Algorithm:

def __init__(self):
        self.ubm      = None
        self.features = []

def process(self, inputs, outputs):

# retrieve the UBM once
        if self.ubm is None:
            self.ubm = gmm_from_data(inputs['ubm'].data)

features = inputs["features"].data.value

stats = bob.machine.GMMStats(self.ubm.dim_c, self.ubm.dim_d)
        self.ubm.acc_statistics(features, stats)

# outputs data
        outputs["statistics"].write({
            't':       numpy.uint64(stats.t),
            'n':       stats.n,
            'sum_px':  stats.sum_px,
            'sum_pxx': stats.sum_pxx,
        })

return True

The code for this algorithm in Python
The ruler at 80 columns indicate suggested POSIX line breaks (for readability).
The editor will automatically enlarge to accomodate the entirety of your input
Use keyboard shortcuts for search/replace and faster editing. For example, use Ctrl-F (PC) or Cmd-F (Mac) to search through this box

For a given set of feature vectors and a Gaussian Mixture Model (GMM), this algorithm computes the 0_th, 1_st and 2_nd order GMM Statistics (Baum-Welch) relying on Bob implementation.

This algorithm relies on the Bob library.

Different of the gmm_statistics algorithm, this one works with only one synchronization channel.

The inputs are:

features: A set of floating point vectors as a two-dimensional array (64 bits) of a client. The number of rows correspond to the number of samples, and the number of columns to the dimensionality of the samples.
ubm: A GMM corresponding to the Universal Background Model.

The output are the statistics of the GMM of a given set of feature vectors (MAP adaptation).

Experiments

Attestation:

Privacy:

Status:

Name	Databases/Protocols	Analyzers
smarcel/tpereira/full_isv_multi/2/btas2015_face-periocular_mobio-female_det	mobio/1@female	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv/2/btas2015_periocular_mobio-female_det_bobv2-0	mobio/1@female	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv_multi/2/btas2015_face-periocular_cpqd-smartphone-male_det	cpqd/1@smartphone_male	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv/2/btas2015_periocular_cpqd-smartphone-male_det	cpqd/1@smartphone_male	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv/2/btas2015_face_cpqd-smartphone-male_det	cpqd/1@smartphone_male	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv_multi/2/btas2015_face-periocular_mobio-male_det	mobio/1@male	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv/2/btas2015_periocular_mobio-male_det	mobio/1@male	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv/2/btas2015_face_mobio-male_det	mobio/1@male	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv_multi/2/btas2015_face-periocular_cpqd-smartphone-female_det	cpqd/1@smartphone_female	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv/2/btas2015_periocular_cpqd-smartphone-female_det	cpqd/1@smartphone_female	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv/2/btas2015_face_cpqd-smartphone-female_det	cpqd/1@smartphone_female	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv_multi/2/btas2015_face-periocular_mobio-female_det	mobio/1@female	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv/2/btas2015_periocular_mobio-female_det	mobio/1@female	tutorial/eerhter_postperf_iso/1
tpereira/tpereira/full_isv/2/btas2015_face_mobio-female_det	mobio/1@female	tutorial/eerhter_postperf_iso/1
martabarrero/smarcel/full_isv/1/Prueba_ISV_2	banca/1@Mc	tutorial/eerhter_postperf/1

This table shows the number of times this algorithm has been successfully run using the given environment. Note this does not provide sufficient information to evaluate if the algorithm will run when submitted to different conditions.

algorithms tpereira gmm_statistics_train 3

Endpoint Groups 1

Unnamed group

Experiments

algorithms

tpereira

gmm_statistics_train

3