.. vim: set fileencoding=utf-8 :
.. Tiago de Freitas Pereira <tiago.pereira@idiap.ch>

================================================================
 Heterogeneous Face Recognition as a Session Variability Problem
================================================================


This section contains instructions on how to reproduce the experiments from Chapter 4 **Heterogeneous Face Recognition as a Session Variability Problem**.

Exceptionally, these instructions will cover only the **Thermal** database.
This would avoid this section to be extremely large.
However, the same set of instructions applies to ALL heterogeneous face databases.
To see all the available databases, check::

 $ resources.py --types database
  List of registered databases:

  - bob.bio.htface 1.0.0 @  <experiment>/bob.bio.htface: 
    + casia-nir-vis-2        --> bob.bio.htface.configs.databases.casia_nir_vis: database
    + cuhk-cufs              --> bob.bio.htface.configs.databases.cuhk_cufs: database
    + cuhk-cufsf             --> bob.bio.htface.configs.databases.cuhk_cufsf: database
    + eprip                  --> bob.bio.htface.configs.databases.eprip: database
    + fargo                  --> bob.bio.htface.configs.databases.fargo: database
    + fargo_depth            --> bob.bio.htface.configs.databases.fargo_depth: database
    + ldhf                   --> bob.bio.htface.configs.databases.ldhf: database
    + nivl                   --> bob.bio.htface.configs.databases.nivl: database
    + pola_thermal           --> bob.bio.htface.configs.databases.pola_thermal: database
    + thermal                --> bob.bio.htface.configs.databases.thermal: database



Thermal Experiments
===================

The sequence of experiments in this subsection generates the necessary data that creates Figures 4.11.
This covers the training using DCT coefficients and LBP histograms. 


DCT Coefficients
----------------

The sequence of experiments in this subsection generates the necessary data to generate Figure 4.11 and Table 4.7.

First, the experiments should be generated::

 $ bob bio htface htface_baseline isv_g64_u50 thermal -vv
 $ bob bio htface htface_baseline isv_g128_u50 thermal -vv
 $ bob bio htface htface_baseline isv_g256_u50 thermal -vv
 $ bob bio htface htface_baseline isv_g512_u50 thermal -vv
 $ bob bio htface htface_baseline isv_g1024_u50 thermal -vv

Once the experiments are executed, the plots and metrics can be generated with the following command::


 $ bob htface evaluate_and_squash \
 \
 <experiment_path>/thermal/isv_g64_u50/ \
 <experiment_path>/thermal/isv_g128_u50/ \
 <experiment_path>/thermal/isv_g256_u50/ \
 <experiment_path>/thermal/isv_g512_u50/ \
 <experiment_path>/thermal/isv_g1024_u50/ \
 \
 --legends "64" --legends "128" --legends "256" --legends "512" --legends "1024" \
 --report-name "thermal.pdf" \
 --x-min 1 \
 --title ""


If everything goes alright the following plot should be dumped

.. image:: ./img/chap4/thermal_dct.png
  :scale: 100 %



LBP Histograms
--------------


The sequence of experiments in this subsection generates the necessary data to generate Figure 4.11 and Table 4.7.

First, the experiments should be generated::

 $ bob bio htface htface_baseline isv_g64_u50_LBP thermal -vv
 $ bob bio htface htface_baseline isv_g128_u50_LBP thermal -vv
 $ bob bio htface htface_baseline isv_g256_u50_LBP thermal -vv
 $ bob bio htface htface_baseline isv_g512_u50_LBP thermal -vv
 $ bob bio htface htface_baseline isv_g1024_u50_LBP thermal -vv

Once the experiments are executed, the plots and metrics can be generated with the following command::


 $ bob htface evaluate_and_squash \
 \
 <experiment_path>/thermal/isv_g64_u50_LBP/ \
 <experiment_path>/thermal/isv_g128_u50_LBP/ \
 <experiment_path>/thermal/isv_g256_u50_LBP/ \
 <experiment_path>/thermal/isv_g512_u50_LBP/ \
 <experiment_path>/thermal/isv_g1024_u50_LBP/ \
 \
 --legends "64" --legends "128" --legends "256" --legends "512" --legends "1024" \
 --report-name "thermal_LBP.pdf" \
 --x-min 1 \
 --title ""


If everything goes alright the following plot should be dumped

.. image:: ./img/chap4/thermal_lbp.png
  :scale: 100 %



ISV Intuition
=============

The ISV intuition showed in Figures 4.1  and 4.2 can also be plotted.
This can be generated with the following command::

 $ bob bio htface isv_intuition <OUTPUT_FILE.pdf>
 

If everything goes alright the following plots should be dumped

.. image:: ./img/chap4/ISV_intuition-0.png
  :scale: 70 %
.. image:: ./img/chap4/ISV_intuition-1.png
   :scale: 70 %

.. image:: ./img/chap4/ISV_intuition-2.png
  :scale: 70 %
.. image:: ./img/chap4/ISV_intuition-3.png
   :scale: 70 %


Relevant publications for this Section
=======================================

 - [`pdf1 <http://publications.idiap.ch/downloads/papers/2017/deFreitasPereira_IEEECOMPUTERSOCIETYWORKSHOPONBIOMETRICS_2016.pdf>`_] FREITAS PEREIRA, TIAGO, and SÉBASTIEN MARCEL. "Heterogeneous Face Recognition using InterSession Variability Modelling." Proceedings of the IEEE Conference on Comput er Vision and Pattern Recognition Workshops. 2016.
 
 - [`pdf2 <http://publications.idiap.ch/downloads/papers/2015/deFreitasPereira_BTAS2015_2015.pdf>`_]  FREITAS PEREIRA, TIAGO, and SÉBASTIEN MARCEL. "Periocular biometrics in mobile environment." Biometrics Theory, Applications and Systems (BTAS), 2015 IEEE 7th International Conference on. IEEE, 2015.