HMMDemo User’s Manual.

1.    Main window.

At startup, application tries to locate USB camera and runs it in YUV mode. In success main application window will appear as the following.

Base view

 

 

Camera view

 
 

 

 


Main window is divided into 2 views. Right view usually contents view from camera (“Camera view”). Left view is for image database (“Base view”).

 

Toolbar elements.

* - create new face database. Current base will be saved to disk.

* - open face database (stored in txt file) or open BMP file.

* - specify current face database parameters.

* - add face into face database.

 - delete face from database.

 - zoom face database images.

*  - decimate face database images.

* - pause/resume camera capture.

 - specify camera resolution

 - adjust video quality parameters.

 - select whole image in Camera view

* - train person(s) HMM(s)

* - recognize face, selected in camera view.

 - delete all HMM information for face base.

 - about HMMDemo.

 

Camera view

Initially camera view contents picture from camera.

Button *pauses/resumes capture.

User can specify rectangular image region on camera view using mouse (please begin selecting region from top left corner and finish at right bottom corner to avoid crashes).

Whole image can be selected by button.

Selected rectangular region can be added into database using *button or recognized using * button.

 

Camera view can also content image from BMP file. BMP file can be opened by *button.

 

 

Base view.

When face database is opened “Base view” contents images from this base. Base view can be stated in 2 modes: “all mode” and “person mode”.

 

In “all mode”, for every person, stored in database, only one picture is shown. This situation is show below.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 In “person mode” all faces of one person are shown. This is depicted below.

 

 

How to switch between “all mode” and “person mode?

To switch from “all mode” to “person mode” double click on image of person you are interested.

To switch from “person mode” to “all mode” double click on free (not image) space of base view.

2.    Creating face base.

At startup, application creates empty database. By default, new base is placed in untitled_base.txt file. To specify new database file press button and fill fields of following dialog

 

Type filename and some signature in specified fields.

 

Now you can add faces into base.

Select region with new face and press *button. If “base view” is in “person mode”, face will be added to person feces list. If “base view” in “all mode” you can be asked to specify person name. You can choose name from list of existing names (in this case image will join corresponding person images) or type new name (in this case new person with specified name will be created)

You can always delete any image or any person from base. For this purposes click on image or person you want to delete and press  button.

 

When you perform add/delete operations, database saves automatically all your changes.

 

3.    Training/Recognition.

 

When you've collected several persons and several photos present each one, you can train HMMs of every person. You can do it both in “all mode” and “person mode”.

To train HMMs of all persons switch to “all mode” and press * button. All untrained HMMs will be trained ( clock-cursor will be shown during training process ).

To train HMM of concrete person switch to “person mode” and  press * button. HMM of person will be trained and during training segmentation of first image of person will appear.

 

Note, than you add/delete person images, corresponding HMM becomes untrained, and after all changes user must train all untrained HMMs.

 

If all HMMs are trained you can try to recognize somebody.

For this

1)     Select face region in “camera view” (from camera of BMP file)

2)     Press * button.

 

Result of recognition will appear as message box with 3 person names. First name will correspond to most similar person etc. “Base view” will be switched into “all mode” and most similar person will be highlighted with magenta rectangle.

 

4.    Batch recognition.

To demonstrate recognition rate of eHMM technique the batch recognition was implemented in HMMDemo.

For batch recognition you must have 2 face base: one as training base and other as base for recognition. Both bases must be “personally” equal, i.e. base for recognition must consist of persons, which present in training base ( true matching is made by names ).

Here is scenario of batch recognition.

1)     Create first face base in usual way. Save it.

2)     Create second base, which consists of people from first base (names must be equal in both bases), using another images. Save it.

3)     Load first base and train it.

4)     Go menu item “Options->Recognize another base”. You will be prompted to define .txt file, which is file of base you want recognize. Choose the base and recognition starts.

5)     Recognition can take much time with large base. Be patient during cursor is “clock”. In the end of recognition massage box will appear. It will report how many images were processed and successfully recognized.

5.    HMMDemo settings

There are several types of settings in HMMDemo concerning use of HMMs.

To obtain settings property sheet go to Options->Settings item of menu.

 

Presently we have the following types of parameters

1)     Sampling parameters

 

2)     HMM parameters

Number of checked checkboxes = number of superstates of HMM.

Value in editbox near checkbox is number of states within corresponding superstate

 

All internal states of HMM have the same number of component of Gaussian mixture (here 3)

 

3)     Other parameters

 

In HMM face recognition it is important that  face images must have not very different sizes. So we must scale images, when we train or recognize. Above dialog allows to specify what scaling will be used.

If “Width” box is checked and “Height” is not checked, then all images will be scaled to have specified width, height of every image will be scaled proportionally.

If “Height” box is checked and “Width” is not checked, then all images will be scaled to have specified height, width of every image will be scaled proportionally.

If “Width” box is checked and “Height” is checked, then all images will be scaled to have specified width and specified height.

 

“Suppress intensity” flag allows not considering lowest DCT coefficient. This can help in case of different illumination conditions in training images and tested image.

6.    Working with settings.

1)     When you start/exit application, it loads/stores settings from/into windows registry.

2)     Every time you train base or recognize person settings, which are displayed in property pages, are used.

3)     If any person of facebase was trained and you save facebase – configuration file, which stores all settings, are saved.

            If the name of facebase file is “Base.txt”, then config file will have “BaseCFG.txt”   name.

If there are no trained persons in the base – config file will not be created.

4)   When you load facebase from file and there is corresp. Config file, then settings from this file will be loaded.

 

5) If you have facebase loaded into application and you changes any parameters from property pages (HMM topology or sampling params) all (trained?) information will be removed. And you must train HMMs again.

 

Also you can store and load settings to/from any other config file.

Use “File->Save config file” and “File->Load config file” for these purposes.

 

Every time you change settings via property pages or loading config file and these new settings differs from you had before – all training information (hmms) are cleared.