The released code is for academyic use purpose only, and WITHOUT ANY
WARRANTY; For any commercial use, please contact HanHu.CAS@gmail.com.
The LTV method demostrated here used a Total Variation package provided by Prof. Wotao Yin (http://www.caam.rice.edu/~optimization/L1/2007/09/software_08.html, GNU General Public License)
***************
To get the details of the 13 methods, please refer to our paper:
Hu Han, Shiguang Shan, Xilin Chen, and Wen Gao. A Comparative Study on Illumination Preprocessing in Face Recognition. Pattern Recognition (P.R.), vol. 46, no. 6, pp. 1691-1699, Jun. 2013.
Please cite the above paper, if our code is used in your research.
***************
Keywords: Face illumination preprocessing, face illumination normalizatioin, lighting normalizaition
How to run?
1¡¢Before running the code, you should get TT package from http://lear.inrialpes.fr/people/triggs/src/amfg07-demo-v1.tar.gz, and unzip add preproc2.m, gauss.m and gaussianfilter.m into the 'IlluminationPreprocessing' directory.
2¡¢Run 'IlluminationNormalization4OneImg.m' and you can get 13 different images produced by 13 vary methods of illumination preprocessing.
3¡¢The parameters of some methods should be optimized based on your face image size. We used parameters for 64*80 face images. We align the face images based on the two-eye centers.
没有合适的资源?快使用搜索试试~ 我知道了~
温馨提示
光照预处理是人脸识中种有效的处理光照变化的方法。近年来涌现出一系列人脸光照预处理方法,但针对这些方法的系统对比与分析的工作相对较少,本文在系统比较现有方法的基础上提出了人脸光照预处理方法的新见解和结论,以及如何设计更好的预处理方法。我们对12种具有代表性的光照预处理方法进行比较研究(HE,LT,GIC,DGD,LOG,SSR,GHP,SQI,LDCT,LTV,LN和TT),着重于两个 新的角度:(1)全局方法的局部化和(2)大尺度和小尺度特征带的融合。在公开的人脸数据库(Yalebext,CMU-PIE,CAS-PEAL和FRGC v2.0)上的实验表明,对全局的光照处理方法(HE,GIC,LTV和TT)进行局部化进一步提高了性能。对(SSR,GHP,SQI,LDCT,LTV和TT)等方法进行大尺度和小尺度的融合有助于光照不变的人脸识别。 来源:http://valser.org/forum.php?mod=viewthread&tid=1051&page=1&extra=#pid1254
资源推荐
资源详情
资源评论
收起资源包目录
FaceIlluminationProcessing_release_1.0.zip (98个子文件)
README.txt 1KB
IlluminationPreprocessing_release_v1.0
LTV.m 1KB
IlluminationNormalization4OneImg.m 7KB
TVL1_v2.32
compile_display.m 1KB
compile_no_display.m 1KB
Changes.txt 291B
Graph_anisoTV_L2_v2_consistent_weights.m 3KB
compile_display_debug.m 1KB
Supported Environments.txt 137B
Readme.txt 782B
Graph_anisoTV_L2_v2.m 5KB
Graph_anisoTV_L1_v2_consistent_weights.m 4KB
Graph_anisoTV_L1_v2.m 6KB
private
mt_TV_L1_1bit_A.mexw32 22KB
mt_TV_L1_16bit.mexw32 21KB
mt_TV_L1_16bit_A.mexw32 22KB
mt_TV_L1_16bit.mexw64 26KB
2009-10-04 mexw32 compiled with MATLAB R2007b MinGW XP 32bit, DEBUG version.txt 0B
dnc_TV_L2_16bit_A.mexw64 24KB
dnc_TV_L2_16bit.mexw64 22KB
dnc_TV_L2_8bit_A.mexw32 20KB
mt_TV_L1_8bit.mexw64 26KB
dnc_TV_L2_16bit_A.mexw32 20KB
mt_TV_L1_1bit.mexw32 21KB
mt_TV_L1_1bit.mexw64 26KB
mt_TV_L1_16bit_A.mexw64 27KB
dnc_TV_L2_1bit_A.mexw32 20KB
dnc_TV_L2_8bit.mexw64 22KB
mt_TV_L1_8bit_A.mexw32 22KB
dnc_TV_L2_1bit_A.mexw64 24KB
dnc_TV_L2_8bit_A.mexw64 24KB
dnc_TV_L2_16bit.mexw32 19KB
mt_TV_L1_8bit.mexw32 21KB
mt_TV_L1_1bit_A.mexw64 28KB
dnc_TV_L2_1bit.mexw32 19KB
dnc_TV_L2_1bit.mexw64 22KB
dnc_TV_L2_8bit.mexw32 19KB
mt_TV_L1_8bit_A.mexw64 27KB
sources
DNC_algorithms
dnc_v2_globalvars.h 7KB
dnc_v2_TV_L2_core.h 13KB
dnc_v2_TVPara.cpp 6KB
MATLAB_compiling_commands.txt 593B
MultiThread_algorithms
MATLAB_compiling_commands.txt 670B
mtl1_TVPara_v2.cpp 6KB
To do's.txt 62B
mtl1_Core_2D_v2.h 23KB
mtl1_globalvars_v2.h 10KB
shared
Common_DNC_MT.h 21KB
Common_DNC_MT_types.h 2KB
OverlappedBlock.m 5KB
retinex_mccann99.m 5KB
DCTN.m 3KB
HE.m 2KB
LoG.m 761B
SQI.m 2KB
GIC_Canonical_YaleBExt_64x80.bmp 6KB
DY.m 595B
GIC_Canonical_PIEIllum_64x80.bmp 6KB
DX.m 595B
ImagesForDebug
00001_P00A-095E+00_GIC.bmp 6KB
00003_P00A+020E+10.bmp 6KB
00001_P00A-050E+00.bmp 6KB
00001_P00A-095E+00.bmp 6KB
00001_P00A-005E-10.bmp 6KB
00001_P00A-035E-20.bmp 6KB
004_105x120.bmp 14KB
00003_P00A+000E+00.bmp 6KB
02463d505_ORI.bmp 6KB
00001_P00A+000E+00.bmp 6KB
001_105x120.bmp 14KB
00001_P00A-070E+00.bmp 6KB
00ORI_02463d683.bmp 14KB
00001_P00A-005E-10_GIC.bmp 6KB
006_105x120.bmp 14KB
04059_27_18.bmp 6KB
003_105x120.bmp 14KB
00002_P00A-060E-20.bmp 6KB
000_105x120.bmp 14KB
I.bmp 6KB
00001_P00A-035E-20_GIC.bmp 6KB
04055_27_09.bmp 6KB
00006_P00A-020E+10.bmp 6KB
I_yaleB.bmp 22KB
00039_P00A-110E-20.bmp 6KB
005_105x120.bmp 14KB
002_105x120.bmp 14KB
00007_P00A+000E+00.bmp 6KB
02463d475_ORI.bmp 6KB
00001_P00A-010E-20.bmp 6KB
GIC_Canonical_YaleBExt_64x80.mat 5KB
retinex_frankle_mccann.m 3KB
RETINEX.m 2KB
GHP.m 828B
GIC.m 5KB
README.txt 1KB
GIC_Canonical_PIEIllum_64x80.mat 4KB
LN.m 3KB
LT.m 606B
共 98 条
- 1
资源评论
- 徐颜卿2018-01-15挺好的资源但是缺少一部分
- fang3582291662016-11-14很强的资源
- Ajing02162016-12-20很好,实用,很全面。
honeyhhan
- 粉丝: 3
- 资源: 1
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功