Coder Social home page Coder Social logo

afs's Introduction

Image

Automatic Feature Selection is a code framework for the feature selection method introduced in

Caner Hazirbas, Julia Diebold, Daniel Cremers, Optimizing the Relevance-Redundancy Tradeoff for Efficient Semantic Segmentation, In Scale Space and Variational Methods in Computer Vision, 2015.

Framework Structure

--AFS    
    |
    -- build            ; build directories for CMake and QtCreator projects
    |
    -- cuda             ; cuda source/header files
    |
    -- doc              ; code documentation and related conference paper
        |
        -- html         ; html documentation (double-click on **index.html**)
        |
        -- latex        ; PDF documentation             
    |
    -- include          ; header files
    |
    -- misc             ; miscellaneous source files and mrmr
        |
        -- mrmr         ; folder to store mrmr executable file
    |
    -- src              ; source files
    |
    -- AFS.pro          ; QtCreator Project File
    |
    -- CMakeLists.txt   ; CMake Project File

How To Build

Required Hardware/Software

To be able to compile and run the code you need a computer with

  • Ubuntu OS (12.04/14.04)
  • NVidia GPU with CUDA support

You need to install the following libraries (I recommend you to download and compile the libraries from source):

Along with these libraries, you need to install CUDA drivers on your machine:

Once libraries are installed, you can download the source from github:

        git clone https://github.com/tum-vision/afs.git

This framework requires mrmr method to be compiled. Please download the source files inside the folder misc/mrmr/ and compile the code with make command.

Build from CMake

  1. Please export the following paths in your .bashrc if you built libraries from source.

    • OpenCV 2.4.10
    export OPENCV_DIR=<path to OpenCV>/OpenCV/2.4.10/share/OpenCV
    export PKG_CONFIG_PATH=$PKG_CONFIG_PATH:<path to OpenCV>/OpenCV/2.4.10/lib/pkgconfig
    export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:<path to OpenCV>/OpenCV/2.4.10/lib  
    
    • Boost 1.54.0
    export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:<path to Boost>/Boost/1.54/build/lib
    
  2. Change the following lines in CMakeLists.txt

    • Line 13 : Set path to the OpenCVConfig.cmake
    • Line 23 : Set CUDA Compute Capability in arch=compute_??,code=sm_??
  3. Run console in AFS/build/CMake and type:

    cmake ../../
    make 
    make install
    make clean (to clean the project)
    

Build from QtCreator(qmake)

  1. Set INCLUDEPATH and LIBS path in AFS.pro (qmake project file) for :

    • OpenCV 2.4.10: Lines 71, 73
    • Boost 1.54.0 : Lines 89, 91
    • CUDA 6.5 : Lines 104, 106
  2. Set CUDA installation directory(CUDA_DIR) and CUDA compute capability(CUDA_ARCH) in Lines 101, 102

  3. You can use build/QtCreator/Debug and build/QtCreator/Release folders to compile the project with QtCreator

Both builds will copy the executable file (AFS) into the main project folder. You can run the executable as ./AFS.

afs's People

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.