/MSRNet

Primary LanguageJupyter Notebook

MSRNet

  • This is the sample code of saliency detection for 2017 cvpr paper [Instance-Level Salient Object Segmentation] by Guanbin Li, Yuan Xie, Liang Lin and Yizhou Yu
  • This code is tested on MATLAB 2014b on Ubuntu14.04
  • For more information, please visit our project page (http://hcp.sysu.edu.cn/instance-level-salient-object-segmentation)

Contents

This code includes

  • 'deeplab-caffe': the Caffe toolbox for Multiscale Refinement Network (MSRNet)
  • 'models_prototxts': pre-trained models and prototxts
  • 'code': codes to do testing
  • 'data':
    • a.imgs: source images to do saliency detection
    • b.pre: predicted results

Usage Instructions

Please follow the instructions below to run the code.

  • Compile the Caffe and matcaffe in the deeplab-caffe package.
  • Put your own images in the data/imgs directory
  • Download the pretrained MSRNet-VGG models by running the script
  ./models_prototxts/get_msrnet-vgg_model.sh
  • Generate saliency map by running the matlab code
  ./code/demo.m

Citation

If you find this useful, please cite our work as follows:

@inproceedings{MSRNet2017object,
  title={Instance-Level Salient Object Segmentation},
  author={Guanbin Li, Yuan Xie, Liang Lin and Yizhou Yu},
  booktitle={CVPR},
  year={2017}
}

Please contact "xiey39@mail2.sysu.edu.cn" if any questions with the code.