ODSAWL

code for "Learning Object Detectors with Semi-Annotated Weak Labels", published in TCSVT2018

Installation MatConvNet and WSDDN

1. Download and install MatConvNet
2. Install this module with the package manager of MatConvNet vl_contrib:

    vl_contrib('install', 'WSDDN') ;
    vl_contrib('setup', 'WSDDN') ;

3. If you want to train a ODSAWL model, download the items below:

   a. PASCAL VOC 2007 devkit and dataset under data folder

   b. Pre-computed edge-boxes and selectiveSearch-boxes for PASCAL VOC 2007 from GoolgeDrive

   c. Pre-trained network from MatConvNet website under model folder

Train and Test

After completing the installation and downloading the required files, you can train and test ODSAWL:

            cd scripts;
            opts.modelPath = '....' ;
            opts.gpu = .... ;
	    opts.labelNumPerCls = ...;
	    opts.iteNum = ...;
            odsawl(opts) ;
                        

gitee

code has also been released in gitee

Citation

@article{zhang2018learning,
  title={Learning Object Detectors With Semi-Annotated Weak Labels},
  author={Zhang, Dingwen and Han, Junwei and Guo, Guangyu and Zhao, Long},
  journal={IEEE Transactions on Circuits and Systems for Video Technology},
  volume={29},
  number={12},
  pages={3622--3635},
  year={2018},
  publisher={IEEE}
}