objectContourDetector

This is the code for arXiv paper Object Contour Detection with a Fully Convolutional Encoder-Decoder Network by Jimei Yang, Brian Price, Scott Cohen, Honglak Lee and Ming-Hsuan Yang, 2016.

Contents

• This code includes

• the Caffe toolbox for Convolutional Encoder-Decoder Networks (caffe-cedn)
• scripts for training and testing the PASCAL object contour detector, and
• scripts to refine segmentation anntations based on dense CRF.

• It is tested on Linux (Ubuntu 14.04) with NVIDIA TITAN X GPU.

Please follow the instructions below to run the code.

Compilation

• Compile the Caffe, matcaffe and pycaffe in the caffe-cedn package.

Training on PASCAL

• Download the pre-processed dataset by running the script

./data/PASCAL/get_pascal_training_data.sh

• Download the VGG16 net for initialization by running the script

./models/get_vgg16_net.sh

• Start training by running the script

./code/train.sh

• Test the learned network by running the script

./code/test.sh

Testing the pre-trained model

• Download the pre-trained model by running the script

./models/PASCAL/get_pretrained_pascal_net.sh

Citation

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

@inproceedings{yang2016object,
  title={Object Contour Detection with a Fully Convolutional Encoder-Decoder Network},
  author={Yang, Jimei and Price, Brian and Cohen, Scott and Lee, Honglak and Yang, Ming-Hsuan},
  journal={arXiv preprint arXiv:1603.04530},
  year={2016}
}

Please contact "jimyang@adobe.com" if any questions.