This is my version of the Fastest Pedestrian Detector in the West
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
CMakeLists.txt arm compatibility Feb 13, 2017
LICENSE fix typo in README; pedetrian -> pedestrian Jul 23, 2017

The Fastest Pedestrian Detector in the West (FPDW)

This is my version of the Fastest Pedestrian Detector in the West based on the paper:

  • P. Dollár, S. Belongie and P. Perona "The Fastest Pedestrian Detector in the West", BMVC 2010.

This version implements only the detection part. For the training part you have to use the P. Dollàr toolbox (link).


  • OpenCV
  • OpenMP
  • SSE
  • Boost

How to build

FPDW works under Linux and Mac Os environments. I recommend a so-called out of source build which can be achieved by the following command sequence:

  • mkdir build
  • cd build
  • cmake ../
  • make -j<number-of-cores+1>

##How to use

###MATLAB - Training Put the file savetoxml inside the folder toolbox/detector. In the file acfDemoInria, after the line:

detector = acfTrainIntegral( opts );

add this code:

filename = 'models/inria_detector.xml';
isTrained = exist(filename,'file');
if( ~isTrained )
    savetoxml(detector, 'models/inria_detector.xml');

###C++ - Detection Go to the bin diretory and launch the program with the following command:

./fpdw ../detector/inria_detector.xml /path/to/the/image.jpg

The detector is already trained with the complete INRIA dataset. So, you can directly use the xml file inside the detector folder.