## LMR: Learning A Two-class Classifier for Mismatch Removal

## Usage:

 Prerequisites: matlab (test on 2015b and 2016b), vlfeat
 
 There are several trained classifiers in "Trained_Model" generated by the authors using the "trainingData".
 1.  Run the LMR_Test.m for mismatch removal using trained classifiers, three models you can choose including:1)Neural Network (Net.mat);  2)Random Forest (RF.mat)  3)SVM (svmStructure.mat)
 2.  Run the LMR_Train.m if you want to train the cllasifier by yourself.
 
Note: if any error exists in vlfeat, please redownload it from its homepage www.vlfeat.org/  ,  then uncompress and rename it.

## Please cite this paper if you use the code in this repository as part of a published research project.
@article{ma2019lmr,
  title={LMR: Learning A Two-class Classifier for Mismatch Removal},
  author={Ma, Jiayi and Jiang, Xingyu and Jiang, Junjun and Zhao, Ji and Guo, Xiaojie},
  journal={IEEE Transactions on Image Processing},
  year={2019},
  publisher={IEEE}
}