Multi-class object detection with BAdaCost.

This repo has a modified version of Piotr Dollar toolbox (Matlab and C++ code) to replicate the experiments we made for our Cost-Sensitive Multiclass algorithm paper. If you use this code for your own research, you must reference our journal paper:

• BAdaCost: Multi-class Boosting with Costs. Antonio Fernández-Baldera, José M. Buenaposada, and Luis Baumela. Pattern Recognition, Elsevier. In press, 2018. DOI:10.1016/j.patcog.2018.02.022

Replicate paper experiments or simply use our trained classifiers.

Our modifications to P.Dollar toolbox have only been tested on GNU/Linux Matlab. To replicate paper experiments you have to:

• Clone this repo
  • From Matlab execute addpath(genpath(PATH_TO_TOOLBOX))
  • From Matlab execute toolboxCompile
• Clone the multi-view car detection scripts repo and follow instructions there.
• Clone the multi-view face detection scripts repo and follow instructions there.