Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How to build the library and use it to train, predict, and get feature from image #1

Open
cia1099 opened this issue Nov 9, 2018 · 3 comments

Comments

@cia1099
Copy link

cia1099 commented Nov 9, 2018

Hello,
I clone this library and use cmake to build, but there are much error.

Can you instruct how to build the code and use this code function example?

I am sincerely grateful if any help of your providing.

@belltailjp
Copy link
Owner

belltailjp commented Nov 9, 2018

Unfortunately I haven't been using this code long time, thus I can provide no support anymore.

My general advice is that you'd better report at least following information. (Please be noted that this doesn't mean I can promise you some answers by responding to these items)

  • In which environment you have tried to build
  • How you ran build process
  • What exact error you've got.

@cia1099
Copy link
Author

cia1099 commented Nov 14, 2018

Hi,

  • I use Visual Studio2012 to complier with OpenCV320 in Win32 environment.

  • I use cmake gui to include CMakeLists folder path to configuration the generation the project, I ran the building process.
    e.g.

  1. I git clone this code on my D:/Project/CoHOG/libcohog/, in this path there is a CMakeLists.txt, I create a folder D:/Project/CoHOG/libcohog/build/ for generation target destination. I canceled the WITH_CUDA and WITH_OMP selection due to make it simplest.
    image
    image

While the cmake generated done, I open the project with VS2012 to build ALL_BUILD.vcxproj, then there are the error as follow figure: (these underline are error)
image

@cia1099
Copy link
Author

cia1099 commented Nov 23, 2018

Hello
I have successfully build the code and run it to get feature.
But I have a problem on how to stack these feature to train.

My development environment is OpenCV320, I want to train INRIA dataset there are 2716 and 12180 for positive and negative respectively. Every data is 64x128 pixels, I divide the size into 4x8 regions. Thus the dimension of feature is 63488 on the INRIA dataset, if I stack the feature with total training data, there are 63488x(2716+12180) size of Mat. Of cause, it is out of memory to stack at such a large Mat.

Can you reply me how CoHog do train data? Thanks

Best regards

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants