---------------------------------------------------------------------------------
Copyright (c) 2013 Xiang Yu, yuxiang03@gmail.com
---------------------------------------------------------------------------------
The code is released with the ICCV 2013 paper in [1]. Further improvements are
quite welcome. It is for research purpose only. Any merchandizing is prohibited.
If you want to re-ensemble or improve the code, please contact the authors in
[1] for permission.
---------------------------------------------------------------------------------
Code content:
"demo.m" is the main entry for running the code.
"face_detect_32" and "face_detect_64" folders contain the inferencing code
(detecting code) from Zhu and Ramanan [3], we would thank them for sharing the
code. Training code for the optimized model as described in the paper [1] is not
included in this demo.
"mex_32" and "mex_64" are mex wrappers of viola-jones face detector, feature
extraction and matching.
"test" folder contains images selected from multiple public databases.
"Model" folder are the model parameters for face detecting. The original Point
Distribution Model parameters have been provided by Saragih et al. [2] that was
trained on Multi-PIE database. We would thank them for providing the resource.
Besides, we added real 3D shape model trained from 3DFE database [4] in order
to solve the pose problem.
The code is provided as is. The authors in [1] have no responsibility for any
debugging, improvement or environment setup.
----------------------------------------------------------------------------------
Landmark initialization ensembling:
In order to robustly provide the initial landmarks for further fitting, we not
only provide our optimized part mixture model (group_sparse_profile) but also
ensembled Zhu-Ramanan detector(P146 model, detect_additional) [3] and viola-jones
face detector to accomplish the task.
----------------------------------------------------------------------------------
Acknowledgement:
We sincerely thank authors in [2] and [3] for their sharing of resource in
developing the code. We also thank authors in [4] for the providing of 3D face
database.
----------------------------------------------------------------------------------
References:
[1] X.Yu, J. Huang, S. Zhang, W. Yan and D.N. Metaxas,
Pose-free Facial Landmark Fitting via Optimized Part Mixtures and Cascaded
Deformable Shape Model.In ICCV 2013.
[2] J. Saragih, S. Lucey and J. Cohn,
Deformable Model Fitting by Regularized Landmark Mean-Shift. In IJCV 2010.
[3] X. Zhu and D. Ramanan.
Face detection, pose estimation and landmark localization in the wild. In CVPR
2012.
[4] Lijun Yin, Xiaozhou Wei, Yi Sun, Jun Wang and Matthew J. Rosato,
A 3D Facial Expression Database For Facial Behavior Research.
7th International Conference on Automatic Face and Gesture Recognition (FGR06),
10-12 April 2006 P:211 - 216
----------------------------------------------------------------------------------
没有合适的资源?快使用搜索试试~ 我知道了~
温馨提示
G. Tzimiropoulos, and M. Pantic, "Optimization problems for fast AAM fitting in-the-wild," ICCV 2013 1. Download the Zhu and Ramanan detector from http://www.ics.uci.edu/~xzhu/face/ 2. Get it to work 3. Re-order the points returned by the detector according to the provided AAM configuration (i.e. the points in Zhu and Ramanan and the points in the provided AAM code are ordered differently) 4. Use the procrustes Matlab built-in function between the detected points (from Zhu and Ramanan) and the mean shape (from AAM).
资源推荐
资源详情
资源评论
收起资源包目录
face_align_iccv_1.1.zip (71个子文件)
face_align_iccv
drawLine.p 519B
activeContour.p 3KB
model
group_sparse_profile.mat 408KB
model_param.mat 495KB
detect_additional.mat 900KB
haarcascade_frontalface_alt2.xml 841KB
LandmarkFitting.p 3KB
face_detect_32
pthreadVC2.dll 55KB
featpyramid.m 2KB
clipboxes.m 297B
features.mexw32 11KB
fconv.mexw32 9KB
pthreadVCE2.dll 61KB
cxcore210.dll 2.1MB
pthreadGCE2.dll 119KB
pthreadVSE2.dll 56KB
reduce.mexw32 9KB
shiftdt.mexw32 9KB
cv210.dll 1.99MB
nms_face.m 2KB
detect.m 5KB
pthreadGC2.dll 117KB
resize.mexw32 9KB
fconvMT.mexw32 9KB
drawLine.m 2KB
README.txt 3KB
face_detect_64
pthreadVC2.dll 81KB
featpyramid.m 2KB
fconv.mexw64 10KB
resize.mexw64 10KB
reduce.mexw64 10KB
clipboxes.m 297B
cxcore210.dll 3.28MB
fconvMT.mexw64 10KB
shiftdt.mexw64 9KB
features.mexw64 12KB
cv210.dll 3.17MB
nms_face.m 2KB
detect.m 5KB
pthreadGC2.dll 182KB
demo.m 2KB
faceAlign.m 2KB
test
4.jpg 26KB
11.jpg 36KB
14.jpg 88KB
12.jpg 84KB
15.jpg 83KB
13.jpg 13KB
3.jpg 32KB
6.jpg 177KB
10.jpg 132KB
1.jpg 18KB
7.jpg 43KB
9.jpg 181KB
2.jpg 61KB
002.png 130KB
8.jpg 112KB
5.jpg 6KB
faceAlign.p 5KB
mex_64
getGradient.mexw64 11KB
FaceDetect.mexw64 9KB
matchTemplate.mexw64 10KB
cxcore210.dll 3.28MB
getLBP.mexw64 12KB
cv210.dll 3.17MB
mex_32
getLBP.mexw32 14KB
FaceDetect.mexw32 9KB
getGradient.mexw32 13KB
matchTemplate.mexw32 13KB
cxcore210.dll 2.1MB
cv210.dll 1.99MB
共 71 条
- 1
资源评论
laughter44
- 粉丝: 0
- 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