wg-perception / partsbaseddetector Goto Github PK

I build the code for Ramanan's Face Detection in c++ implementation ,Also included the Dependencies files like.

1. Boost Python
2. cvmatio

Whether the cvmatio dependencies are needed or not?.

Which model xml I have to use for detecting the faces?.

I used arguments to give the inputs

argv[1] = Xml file path. ---> Person_8parts.xml
argv[2] = image path. ---> input_1.jpg
argv[3] = image path. ---> input_2.jpg

If any one tested the code in c++.Please guide me

It seems that in OpenCV 3.0 FilterEngine does not exist as public class (opencv/opencv#4819). So PartBasedDetector library can't be compiled since it uses that class.

Sabia que se tomar café exageradamente pode fazer mal a saudade?
Sim por isso temos que tomar café moderado e sem exceção.

Not sure what is causing this. Is it on my side or a problem with the build?
Can/Should I specify specific openCv?

When I run the 'object_recognition_by_parts' executable in the devel/lib folder from the terminal with the any of the person.xml model provided and a random image, I get an 'error deserializing file'. Can anybody please help me out?

using the software as is (PartsBasedDetector-master) I get the error
CMake Error at src/CMakeLists.txt:36 (install):
install Library TARGETS given no DESTINATION!

I'm using windows xp and cmake 2.8.10.2 and vs2008

NOTE:
If I remove the "LIBRARY" keyword at the line 36 than it passes the cmake but I don't get a lib project only dll.

Hope you can send me solution.

Hey all,
Can you help me out a bit. I am trying to run this project and am following the given cmake routine but i am facing following error-

CMake Error: The following variables are used in this project, but they are set to NOTFOUND.
Please set them or make sure they are set and tested correctly in the CMake files:
Boost_INCLUDE_DIR (ADVANCED)
used as include directory in directory /var/www/ccpp/PartsBasedDetector-master
used as include directory in directory /var/www/ccpp/PartsBasedDetector-master/src

-- Configuring incomplete, errors occurred!

I've also reinstalled Boost library but it is still the same.
Basically I want get this project running but I am not able to do that. Can anyone help with an overview.

Any help is much appreciated.

Thanks !

I can't seem to find Part.hpp anywhere. It's being included in Part.cpp. Does anyone have it?

Hi Vincent,

     I downloaded the source in the github Link: https://github.com/wgperception/PartsBasedDetector.

Also i followed the README file and done the Steps to build the code.

My Testing Environment:

1. ubuntu 14.0 LTS
2. gcc

Step 1: Downloaded the Zip, Extract it to my Local Directory.

Step 2: Then,build the boost by running the script bootstrap.sh.

Step 3: Downloaded the cvamtio file from the link: https://github.com/hbristow/cvmatio also used this package too.

Step 4: Then I build the code using cmakelist and created the .exe, But no instruction's is provided to give input's.

Step 5: Downloaded the Modal files from the link: https://github.com/wg-perception/PartsBasedDetectorModels. The face xml's are not working for me.

Step 6: Generated .exe by using cmake.

Step 7: Then, I tested the code by giving all the xml files one by one.Finally the Person_8parts.xml is working, but not predicted properly.

Step 8. I gave input format like this:
./partsbaseddetector person_8parts.xml test.jpg test.jpg

     ./partsbaseddetector    - .exe name
     person_8parts.xml       -  argv[1]
     test.jpg                       -  argv[2]
     test.jpg                       -  argv[3]

I have to know,whether i did the correct steps or not.

I am testing this with a bunch of images with the Face_small_146filters.xml but am getting poor results.

However, the same images work well with the original matlab code at http://www.ics.uci.edu/~xzhu/face/ (face_p146_small.mat). I am not sure if the difference is in the models or the implementation.

Has anyone encountered this before?

Sample test: http://www.pwc.com/us/en/careers/campus/assets/img/homepage/hero-people.jpg

I have compiled PartsBasedDetector without MatlabIO. This can only read models with xml or yaml format. The model training_demo.m code (which I run in octave) saves the model as mat format.

model = trainmodel(name,pos,neg,K,pa,sbin);
save('Demo_model.mat', 'model', 'pa', 'sbin', 'name');

There does not seem to be a clear way to covert from .mat to .xml, or am I missing something?

Has matio been considered to read models in C++? http://sourceforge.net/projects/matio/

Regards,

Tim

I am having problems with installation on Windows 7. I am using the CMake GUI and I get the following errors. Apparently there are problems with Point Cloud Library. I haven't intstalled it. Do I need it for this application?

CMake Error at C:/Program Files (x86)/CMake 2.8/share/cmake-2.8/Modules/FindBoost.cmake:1192 (message):
Unable to find the requested Boost libraries.

Unable to find the Boost header files. Please set BOOST_ROOT to the root
directory containing Boost or BOOST_INCLUDEDIR to the directory containing
Boost's headers.
Call Stack (most recent call first):
CMakeLists.txt:86 (find_package)

CMake Error at CMakeLists.txt:88 (find_package):
By not providing "FindPCL.cmake" in CMAKE_MODULE_PATH this project has
asked CMake to find a package configuration file provided by "PCL", but
CMake did not find one.

Could not find a package configuration file provided by "PCL" with any of
the following names:

PCLConfig.cmake
pcl-config.cmake

Add the installation prefix of "PCL" to CMAKE_PREFIX_PATH or set "PCL_DIR"
to a directory containing one of the above files. If "PCL" provides a
separate development package or SDK, be sure it has been installed.

Configuring incomplete, errors occurred!

hi, I build the project in Mac. when I do 'make', some problems occur like this:

$ make
Scanning dependencies of target PartsBasedDetector_lib
[ 6%] Building CXX object src/CMakeFiles/PartsBasedDetector_lib.dir/DepthConsistency.cpp.o
clang: error: no such file or directory: 'NOTFOUND'
make[2]: *** [src/CMakeFiles/PartsBasedDetector_lib.dir/DepthConsistency.cpp.o] Error 1
make[1]: *** [src/CMakeFiles/PartsBasedDetector_lib.dir/all] Error 2
make: *** [all] Error 2

could you help me? thx

Hi, I use the matlab code of this pose estimation paper. It's about 2 seconds to parse an image with resolution of 150x100, which only contains one person. Can you tell me the speed of your code?

I am trying to compile the project. I ve got two options using gcc for Ubutnu 12.04 I got the same issue as the closed issue #25. I ve isntalled doxygen but the error remain the same. The second option from windows xp but I not know which compiler to use. I ve installed msvc. The better option is to compile it from ubuntu but how can I fix the error?

make[2]: DOXYGEN_EXECUTABLE-NOTFOUND: Command not found
make[2]: *** [CMakeFiles/PartsBasedDetector_doc] Error 127
make[1]: *** [CMakeFiles/PartsBasedDetector_doc.dir/all] Error 2
make: *** [all] Error 2

My guess was that doxygen was optional.

Maybe I am doing something wrong... but I am getting the following error:

Candidate.hpp:197:28: error: no viable conversion from 'cv::Matx<float, 3, 1>' to 'const cv::Mat'

Would it be possible to add more comments to training_demo.m that clarify the role of the K and pa arrays, in a way that someone with a math and software background (but not a computer vision background) could understand?

I get the following error during CMake

CMake Error at CMakeLists.txt:89 (find_package):
By not providing "FindEigen.cmake" in CMAKE_MODULE_PATH this project has
asked CMake to find a package configuration file provided by "Eigen", but
CMake did not find one.

I have already installed libeingen2 and libeigen3. Any ideas?

Thanks

When I try to build the project using VS2010, there are many build errors. After I dig into the code, I found the definition of the class 'Parts' is defined in the 'parts.hpp', but in the implementation file 'part.cpp', the class name changed to 'Part'. First I thought it's just a typo, but after fixing to keep the class name consistent, errors still exist. Has anyone met this problem before?

I tried running the detection in c++ and matlab but it is not the same. The matlab detection is better than the c++ one. What may be the reason? What are the difference between the matlab and the c++ code? I am using the Person_26parts model.

How can i obtain the pose estimation for a given face?
I understand that has to be with the best mixture, maybe?
Would aprecciate a little help, ty

I am trying to run demo.cpp which is located in the src folder. Two arguments need to be passed : (1) The model file and (2) The image file

I assume the image file is the one in which we want to detect objects. I was wondering where a sample pre-trained model file can be located.

Can someone point me to it?

Thanks,

Sid

Hello, I wonder if any of these given models is suitable for human pose estimation? I tried out Person_26, Person_8 and PersonINRIA, but none of them works. Can anyone point me the right model, or do I need to train one?

I've been reading the documentation and trying to work out how to extract the position of the parts of the top pose but to no avail.

How do I output the x y positions of each part in the model?

I am getting error while building the code with cvmatio. I have separately downloaded the module form github link. build it. then tried building this code. Still I am not getting it build with cvmatio. I think I need to link it with system library. I am new with linux. Can anyone help me out with the building and linking process..?

I was successful in compiling the binary for windows (had to add export macros for the classes to generate the lib file).

But when I run the exe with a simple model, it runs until the convolution (and prints a convolution time) and then crashes with an error like this -

OpenCV Error: Insufficient memory (Failed to allocate 64244 bytes) in unknown fu
nction, file ......\src\opencv\modules\core\src\alloc.cpp, line 52

It fails with all images (small or large). Has anyone encountered this before?

I am able to run cmake command successfully . When I run make , I get below errors

zayyon@zayyon-Lenovo-ideapad-300-15ISK:~/Downloads/PartsBasedDetector-master/build$ sudo make
[ 0%] Built target PartsBasedDetector_doc
[ 85%] Built target PartsBasedDetector_lib
[ 92%] Linking CXX executable PartsBasedDetector
CMakeFiles/PartsBasedDetector_bin.dir/demo.cpp.o: In function cv::_OutputArray::_OutputArray<float>(cv::Mat_<float>&)': /usr/local/include/opencv2/core/mat.hpp:1149: undefined reference tovtable for cv::_OutputArray'
/usr/local/include/opencv2/core/mat.hpp:1149: undefined reference to vtable for cv::_OutputArray' CMakeFiles/PartsBasedDetector_bin.dir/demo.cpp.o: In functionmain':
/home/zayyon/Downloads/PartsBasedDetector-master/src/demo.cpp:90: undefined reference to cv::imread(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, int)' /home/zayyon/Downloads/PartsBasedDetector-master/src/demo.cpp:96: undefined reference tocv::imread(std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, int)'
libPartsBasedDetector.so: undefined reference to cv::write(cv::FileStorage&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, cv::Mat const&)' libPartsBasedDetector.so: undefined reference tocv::FileNode::operator[](std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&) const'
libPartsBasedDetector.so: undefined reference to cv::imshow(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, cv::_InputArray const&)' libPartsBasedDetector.so: undefined reference tocv::_OutputArray::_OutputArray(cv::Mat&)'
libPartsBasedDetector.so: undefined reference to cv::operator-(cv::MatConstIterator const&, cv::MatConstIterator const&)' libPartsBasedDetector.so: undefined reference tocv::FileNodeIterator::readRaw(std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, unsigned char*, unsigned long)'
libPartsBasedDetector.so: undefined reference to cv::Exception::Exception(int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, int)' libPartsBasedDetector.so: undefined reference tocv::operator<<(cv::FileStorage&, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&)'
libPartsBasedDetector.so: undefined reference to cv::_InputArray::_InputArray(cv::Mat const&)' libPartsBasedDetector.so: undefined reference tocv::write(cv::FileStorage&, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, int)'
libPartsBasedDetector.so: undefined reference to cv::FileStorage::open(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)' libPartsBasedDetector.so: undefined reference tovtable for cv::_InputArray'
libPartsBasedDetector.so: undefined reference to cv::putText(cv::Mat&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, cv::Point_<int>, int, double, cv::Scalar_<double>, int, int, bool)' libPartsBasedDetector.so: undefined reference to cv::split(cv::Mat const&, std::vector<cv::Mat, std::allocatorcv::Mat >&)'
libPartsBasedDetector.so: undefined reference tocv::FileStorage::writeRaw(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned char const_, unsigned long)' libPartsBasedDetector.so: undefined reference to cv::WriteStructContext::WriteStructContext(cv::FileStorage&, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, int, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&)'
libPartsBasedDetector.so: undefined reference tocv::Formatter::get(char const_)' libPartsBasedDetector.so: undefined reference to cv::WriteStructContext::~WriteStructContext()'
libPartsBasedDetector.so: undefined reference tocv::write(cv::FileStorage&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, float)' libPartsBasedDetector.so: undefined reference to cv::namedWindow(std::_cxx11::basic_string<char, std::char_traits, std::allocator > const&, int)'
libPartsBasedDetector.so: undefined reference to`cv::createLinearFilter(int, int, cv::InputArray const&, cv::Point, double, int, int, cv::Scalar const&)'
collect2: error: ld returned 1 exit status
src/CMakeFiles/PartsBasedDetector_bin.dir/build.make:118: recipe for target 'src/PartsBasedDetector' failed
make[2]: *** [src/PartsBasedDetector] Error 1
CMakeFiles/Makefile2:161: recipe for target 'src/CMakeFiles/PartsBasedDetector_bin.dir/all' failed
make[1]: *** [src/CMakeFiles/PartsBasedDetector_bin.dir/all] Error 2
Makefile:127: recipe for target 'all' failed
make: *** [all] Error 2

Hello, I run your code in my computer, and find a bug.
Perhaps the bug is caused by OpenCV.
My OpenCV version is 2.3.
segmentation fault. >> FileStorageModel.cpp:148
As the component-0 part-0 defid is empty. It causes: segmentation fault

I try to run your code in another PC(OpenCV 2.4). It works!

Makes compiling tricky!

I am running opencv 2.4 and getting

Undefined symbols for architecture x86_64: "vtable for FileStorageModel", referenced from:
      FileStorageModel::FileStorageModel()

I believe everything has compiled and I am trying to run the src/demo.cpp and cannot. Any help is appreciated. That was running on a Mac.

On linux, I am getting the following error:

PartsBasedDetector/src/MatlabIOModel.cpp:41:24: fatal error: MatlabIO.hpp: No such file or directory compilation terminated.

On a branch-new clone of 09a2627:

% cd PartsBasedDetector
% mkdir build
% cd build
% cmake ..

CMake Error: Error required internal CMake variable not set, cmake may be not be built correctly.
Missing variable is:
CMAKE_FIND_LIBRARY_PREFIXES
CMake Error: Error required internal CMake variable not set, cmake may be not be built correctly.
Missing variable is:
CMAKE_FIND_LIBRARY_SUFFIXES
CMake Error at /opt/ros/fuerte/share/catkin/cmake/assert.cmake:17 (message):

Assertion failed: check for file existence, but filename
(RT_LIBRARY-NOTFOUND) unset. Message: RT Library

Call Stack (most recent call first):
/opt/ros/fuerte/share/catkin/cmake/tools/rt.cmake:42 (assert_file_exists)
/opt/ros/fuerte/share/catkin/cmake/all.cmake:51 (include)
/opt/ros/fuerte/share/catkin/cmake/catkin-config.cmake:6 (include)
CMakeLists.txt:16 (find_package)

-- Configuring incomplete, errors occurred!

This is on ubuntu 12.04.1.

% cmake --version
cmake version 2.8.7

Thanks!

I am trying to use the modelTransfer to transfer the model from "Face Detection, Pose Estimation and Landmark Localization in the Wild"(http://www.ics.uci.edu/~xzhu/face/) to PartsBasedDetector. However, it seems the it misses most of the faces. I am wondering:
-> is this algorithm equivalent(roughly,within in rounding error etc) to Zhu's mixtures of trees algorithm(after transformed the model)?

Any help is appreciated

When running the makefile I'm getting errors related to opencv, starts with "error: unknown type name 'Scalar'" and a full list of this type, at the file "filter.cpp".
I've tried to include an opencv library at the beginning, and then I'm getting "opencv2/imgproc/hal/hal.hpp' file not found".
So I've tried to change this include to include the file "imgproc.hpp" instead, but then I'm getting a list of errors: "error: use of undeclared identifier 'BaseRowFilter', 'BaseColumnFilter', 'BaseFilter' ..."

Do you have any idea for me?
I'm trying to compile on OSX 10.11 El Capitan with opencv 3.1.0

Thanks