Face Detection Toolbox

Faces detection using Local Binary Patterns and Haar features

Description

Faces detection toolbox v 0.1
------------------------------
This toolbox provides some tools for faces detection using Local Binary Patterns and Haar features.
The task of detection is done by boosting approaches such Adaboosting, FastAdaboosting and Gentleboosting.
The main objective of this toolbox is to deliver simple but efficient tools mainly written in C codes with
a matlab interface and easy to modify.

Installation
------------
This toolbox has been tested on Windows system and should work also on Linux plateform without any problem.

Please run "mexme_fdt" to compile each mex-files and add fdtool directory in the matlab path.

Please open *.c files to have a description of each function and how to use them in Matlab.

Demos
-----
4 demos are included.

i) "demo_mblbp"
ii) "demo_chlbp"
iii) "demo_haar"
vi) "demo_detector_haar"
 
 Main References : [1] R.E Schapire and al "Boosting the margin : A new explanation for the effectiveness of voting methods". The annals of statistics, 1999

                    [2] Zhang, L. and Chu, R.F. and Xiang, S.M. and Liao, S.C. and Li, S.Z, "Face Detection Based on Multi-Block LBP Representation", ICB07

                    [3] C. Huang, H. Ai, Y. Li and S. Lao, "Learning sparse features in granular space for multi-view face detection", FG2006
 
                    [4] P.A Viola and M. Jones, "Robust real-time face detection", International Journal on Computer Vision, 2004

                    [5] M-T. Pham and all, "Detection with multi-exit asymetric boosting", CVPR'08

Greetings to : i ) Ole Jensen for providing me his faces database and the merging detections algorithm for detector_haar and detector_mblbp,
                    ii) and also to Pham Minh Tri for his responses concerning Fastadaboosting and multi-exit asymetric boosting.

Acknowledgements

The author wishes to acknowledge the following in the creation of this submission:
VCAPG2, windows internet explorer + google search = URLs for jpg images

• Required Products: Image Acquisition Toolbox; Image Processing Toolbox
• MATLAB release: MATLAB 7.8 (R2009a)
• Other requirements: A C compiler
• Published M Files: Demo illustrating (Center-Symetric) MultiBlock Local Binary Pattern (mblbp), Demo illustrating Circular Histogram Local Binary Pattern (chlbp), Demo illustrating HAAR features

Other Files

fdtool/0000_-12_0_0_15_0_1.pgm,
fdtool/2.bmp,
fdtool/basicroc.m,
fdtool/build_negatives.m,
fdtool/chlbp.c,
fdtool/chlbp.mexw32,
fdtool/chlbp_adaboost_binary_model_cascade.c,
fdtool/chlbp_adaboost_binary_model_cascade.mexw32,
fdtool/chlbp_adaboost_binary_predict_cascade.c,
fdtool/chlbp_adaboost_binary_predict_cascade.mexw32,
fdtool/chlbp_gentleboost_binary_model_cascade.c,
fdtool/chlbp_gentleboost_binary_model_cascade.mexw32,
fdtool/chlbp_gentleboost_binary_predict_cascade.c,
fdtool/chlbp_gentleboost_binary_predict_cascade.mexw32,
fdtool/class57.jpg,
fdtool/d2uint8_image.m,
fdtool/demo_chlbp.m,
fdtool/demo_detector_haar.m,
fdtool/demo_haar.m,
fdtool/demo_mblbp.m,
fdtool/detection_haar_2P.jpg,
fdtool/detector_haar.c,
fdtool/detector_haar.mexw32,
fdtool/detector_mblbp.c,
fdtool/detector_mblbp.mexw32,
fdtool/display_database.m,
fdtool/eval_chlbp.c,
fdtool/eval_chlbp.mexw32,
fdtool/eval_haar.c,
fdtool/eval_haar.mexw32,
fdtool/eval_haar_subwindow.c,
fdtool/eval_haar_subwindow.mexw32,
fdtool/eval_mblbp.c,
fdtool/eval_mblbp.mexw32,
fdtool/fast_haar_ada_weaklearner.c,
fdtool/fast_haar_ada_weaklearner.mexw32,
fdtool/fast_haar_adaboost_binary_model_cascade.c,
fdtool/fast_haar_adaboost_binary_model_cascade.mexw32,
fdtool/generate_data.m,
fdtool/getmapping.m,
fdtool/gui/coeff_edit_data_callback1.m,
fdtool/gui/coeff_edit_data_callback10.m,
fdtool/gui/coeff_edit_data_callback11.m,
fdtool/gui/coeff_edit_data_callback12.m,
fdtool/gui/coeff_edit_data_callback13.m,
fdtool/gui/coeff_edit_data_callback14.m,
fdtool/gui/coeff_edit_data_callback15.m,
fdtool/gui/coeff_edit_data_callback16.m,
fdtool/gui/coeff_edit_data_callback17.m,
fdtool/gui/coeff_edit_data_callback18.m,
fdtool/gui/coeff_edit_data_callback19.m,
fdtool/gui/coeff_edit_data_callback2.m,
fdtool/gui/coeff_edit_data_callback20.m,
fdtool/gui/coeff_edit_data_callback21.m,
fdtool/gui/coeff_edit_data_callback22.m,
fdtool/gui/coeff_edit_data_callback23.m,
fdtool/gui/coeff_edit_data_callback24.m,
fdtool/gui/coeff_edit_data_callback25.m,
fdtool/gui/coeff_edit_data_callback3.m,
fdtool/gui/coeff_edit_data_callback4.m,
fdtool/gui/coeff_edit_data_callback5.m,
fdtool/gui/coeff_edit_data_callback6.m,
fdtool/gui/coeff_edit_data_callback7.m,
fdtool/gui/coeff_edit_data_callback8.m,
fdtool/gui/coeff_edit_data_callback9.m,
fdtool/gui/coeff_push_callback1.m,
fdtool/gui/config_gui.m,
fdtool/gui/database_edit_callback1.m,
fdtool/gui/database_edit_callback2.m,
fdtool/gui/display_selected_pattern.m,
fdtool/gui/features_list_callback1.m,
fdtool/gui/features_push_add_callback1.m,
fdtool/gui/features_push_remove_callback1.m,
fdtool/gui/gui_features_dictionary.m,
fdtool/gui/gui_menu_open_data_callback.m,
fdtool/gui/gui_menu_save_data_callback.m,
fdtool/gui/gui_number_of_features.m,
fdtool/gui/nbfeat_haar.m,
fdtool/haar.c,
fdtool/haar.mexw32,
fdtool/haar_ada_weaklearner.c,
fdtool/haar_ada_weaklearner.mexw32,
fdtool/haar_adaboost_binary_model_cascade.c,
fdtool/haar_adaboost_binary_model_cascade.mexw32,
fdtool/haar_adaboost_binary_predict_cascade.c,
fdtool/haar_adaboost_binary_predict_cascade.mexw32,
fdtool/haar_dico_19.mat,
fdtool/haar_dico_2.mat,
fdtool/haar_dico_4.mat,
fdtool/haar_dico_5.mat,
fdtool/haar_featlist.c,
fdtool/haar_featlist.mexw32,
fdtool/haar_gentle_weaklearner.c,
fdtool/haar_gentle_weaklearner.mexw32,
fdtool/haar_gentleboost_binary_model_cascade.c,
fdtool/haar_gentleboost_binary_model_cascade.mexw32,
fdtool/haar_gentleboost_binary_predict_cascade.c,
fdtool/haar_gentleboost_binary_predict_cascade.mexw32,
fdtool/Haar_matG.m,
fdtool/haar_scale.c,
fdtool/haar_scale.mexw32,
fdtool/html/demo_chlbp.png,
fdtool/html/demo_chlbp_01.png,
fdtool/html/demo_chlbp_02.png,
fdtool/html/demo_chlbp_03.png,
fdtool/html/demo_chlbp_04.png,
fdtool/html/demo_chlbp_05.png,
fdtool/html/demo_chlbp_06.png,
fdtool/html/demo_chlbp_07.png,
fdtool/html/demo_chlbp_08.png,
fdtool/html/demo_chlbp_09.png,
fdtool/html/demo_chlbp_10.png,
fdtool/html/demo_chlbp_11.png,
fdtool/html/demo_chlbp_12.png,
fdtool/html/demo_chlbp_13.png,
fdtool/html/demo_chlbp_14.png,
fdtool/html/demo_chlbp_15.png,
fdtool/html/demo_chlbp_16.png,
fdtool/html/demo_haar.png,
fdtool/html/demo_haar_01.png,
fdtool/html/demo_haar_02.png,
fdtool/html/demo_haar_03.png,
fdtool/html/demo_haar_04.png,
fdtool/html/demo_haar_05.png,
fdtool/html/demo_haar_06.png,
fdtool/html/demo_haar_07.png,
fdtool/html/demo_haar_08.png,
fdtool/html/demo_haar_09.png,
fdtool/html/demo_haar_10.png,
fdtool/html/demo_haar_11.png,
fdtool/html/demo_haar_12.png,
fdtool/html/demo_haar_13.png,
fdtool/html/demo_haar_14.png,
fdtool/html/demo_haar_15.png,
fdtool/html/demo_haar_16.png,
fdtool/html/demo_mblbp.png,
fdtool/html/demo_mblbp_01.png,
fdtool/html/demo_mblbp_02.png,
fdtool/html/demo_mblbp_03.png,
fdtool/html/demo_mblbp_04.png,
fdtool/html/demo_mblbp_05.png,
fdtool/html/demo_mblbp_06.png,
fdtool/html/demo_mblbp_07.png,
fdtool/html/demo_mblbp_08.png,
fdtool/html/demo_mblbp_09.png,
fdtool/html/demo_mblbp_10.png,
fdtool/html/demo_mblbp_11.png,
fdtool/html/demo_mblbp_12.png,
fdtool/html/demo_mblbp_13.png,
fdtool/html/demo_mblbp_14.png,
fdtool/html/demo_mblbp_15.png,
fdtool/html/demo_mblbp_16.png,
fdtool/html/demo_mblbp_17.png,
fdtool/html/demo_mblbp_18.png,
fdtool/html/demo_mblbp_19.png,
fdtool/ieJPGSearch.m,
fdtool/image_integral_standard.m,
fdtool/image_standard.m,
fdtool/imresize.c,
fdtool/imresize.mexw32,
fdtool/int8tosparse.c,
fdtool/int8tosparse.mexw32,
fdtool/inv_integral_image.m,
fdtool/jensen_24x24.mat,
fdtool/mblbp.c,
fdtool/mblbp.mexw32,
fdtool/mblbp_ada_weaklearner.c,
fdtool/mblbp_ada_weaklearner.mexw32,
fdtool/mblbp_adaboost_binary_model_cascade.c,
fdtool/mblbp_adaboost_binary_model_cascade.mexw32,
fdtool/mblbp_adaboost_binary_predict_cascade.c,
fdtool/mblbp_adaboost_binary_predict_cascade.mexw32,
fdtool/mblbp_featlist.c,
fdtool/mblbp_featlist.mexw32,
fdtool/mblbp_gentle_weaklearner.c,
fdtool/mblbp_gentle_weaklearner.mexw32,
fdtool/mblbp_gentleboost_binary_model_cascade.c,
fdtool/mblbp_gentleboost_binary_model_cascade.mexw32,
fdtool/mblbp_gentleboost_binary_predict_cascade.c,
fdtool/mblbp_gentleboost_binary_predict_cascade.mexw32,
fdtool/mexme_fdt.m,
fdtool/mexopts_intel10.bat,
fdtool/model_detector_chlbp_24x24.mat,
fdtool/model_detector_haar_24x24.mat,
fdtool/model_detector_haar_24x24_nP4.mat,
fdtool/model_detector_mblbp_24x24_4.mat,
fdtool/multi_exit_asymetric_boosting.m,
fdtool/nbfeat_haar.m,
fdtool/negatives/BigTree2.jpg,
fdtool/negatives/face-in-trees-illusion.jpg,
fdtool/negatives/franklin_trees_01.jpg,
fdtool/negatives/glorious-tree-01.jpg,
fdtool/negatives/hollow_tree.jpg,
fdtool/negatives/majestic-trees.jpg,
fdtool/negatives/maple-trees-12.4.jpg,
fdtool/negatives/pic_wonder_crooked_trees_lg.jpg,
fdtool/negatives/Sago_Palm_Trees.jpg,
fdtool/negatives/state_trees.jpg,
fdtool/negatives/strange-trees1.jpg,
fdtool/negatives/trees.jpg,
fdtool/negatives/trees5.jpg,
fdtool/negatives/trees_mem_planes.jpg,
fdtool/negatives/trees_mtns.jpg,
fdtool/negatives/Wawona_Big_Trees_1953.jpg,
fdtool/negatives/wicked-trees.jpg,
fdtool/plot_rectangle.m,
fdtool/readme.txt,
fdtool/rgb2gray.c,
fdtool/rgb2gray.mexw32,
fdtool/train_segment.m,
fdtool/vcapg2.cpp,
fdtool/viola_24x24.mat,
license.txt

[Bodla Rakesh]

 

[Soeren Sproessig]

 

[Soeren Sproessig]

excellent work!

to use it with matlab R2009b x64 on win7pro64 I had to change mexme_fdt.m a little bit:

44c44,46
< lib = ['"' , fullfile(matlabroot , sprintf('extern\\lib\\%s\\microsoft\\libmwblas.lib',computer('arch'))) , '"'];
---
>
> lib = ['"' , fullfile(matlabroot , 'extern\lib\win32\microsoft\libmwblas.lib') , '"'];
>
135c137

[Soeren Sproessig]

in addition to my prior post:
to build vcapg2.cpp you need to have a current Microsoft DirectX SDK installed...

[Soeren Sproessig]

to build vcagp2.cpp (used in demo_detector_haar as camera acquisition provider) there are some other tricks
1. create a dummy dxtrans.h which seems to be missing in the SDK:
c:\Program Files\Microsoft SDKs\Windows\v6.0A\Include\dxtrans.h
#define __IDxtCompositor_INTERFACE_DEFINED__
#define __IDxtAlphaSetter_INTERFACE_DEFINED__
#define __IDxtJpeg_INTERFACE_DEFINED__
#define __IDxtKey_INTERFACE_DEFINED__

see: sample from http://www.riseoftheants.com/mmx/faq.htm#mjpegsample

2. some changes in code to be able to compile with VC08 and the current SDK in Win64:

82d81
<
629,630c628
< // pbmi->hInst = (HINSTANCE)GetWindowLong(GetFocus(),GWL_HINSTANCE);
< pbmi->hInst = (HINSTANCE)GetWindowLongPtr(GetFocus(), GWLP_HINSTANCE);
---
> pbmi->hInst = (HINSTANCE)GetWindowLong(GetFocus(),GWL_HINSTANCE);
958c956
< wc.lpfnWndProc =(WNDPROC) WndProc;
---
> wc.lpfnWndProc = WndProc;

[Sebastien Paris]

Thanks you Soeren for these indications and comments.
Please if you have better models than those already included, I'll include them in a future release.

Sébastien

[Nessrine]

i 'm not soo good in matlab implementation but i would like to test this code. can any one help me specialy i have no idea to use .c file with matlab for me this is the first matter.
thank an advance

[Sebastien Paris]

Add fdtool dir in your path and run "mexme_fdt.m" to compile *.c files