Skip to content
No description, website, or topics provided.
Python Shell
Branch: master
Clone or download
Latest commit a979dbc Jun 15, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data code released Jun 12, 2019
geotnf code released Jun 12, 2019
img add teaser image Jun 13, 2019
model code released Jun 12, 2019
parser code released Jun 12, 2019
trained_models code released Jun 12, 2019
util code released Jun 12, 2019
README.md Update Jun 14, 2019
__init__.py code released Jun 12, 2019
config.py code released Jun 12, 2019
eval.py code released Jun 12, 2019
eval.sh code rlease Jun 12, 2019
requirements.txt code released Jun 12, 2019
train-coseg.py code released Jun 12, 2019
train.py code released Jun 12, 2019
train.sh code released Jun 12, 2019

README.md

Show, Match and Segment: Joint Learning of Semantic Matching and Object Co-segmentation

This repository contains the source code for the paper Show, Match and Segment: Joint Learning of Semantic Matching and Object Co-segmentation.

Citation

If you find our code useful, please consider citing our work using the following bibtex:

@article{MaCoSNet,
    title={Show, Match and Segment: Joint Learning of Semantic Matching and Object Co-segmentation},
    author={Chen, Yun-Chun and Lin, Yen-Yu and Yang, Ming-Hsuan and Huang, Jia-Bin},
    journal={arXiv},
    year={2019}
}

@inproceedings{WeakMatchNet,
  title={Deep Semantic Matching with Foreground Detection and Cycle-Consistency},
  author={Chen, Yun-Chun and Huang, Po-Hsiang and Yu, Li-Yu and Huang, Jia-Bin and Yang, Ming-Hsuan and Lin, Yen-Yu},
  booktitle={Asian Conference on Computer Vision (ACCV)},
  year={2018}
}

Environment

  • Install Anaconda Python3.7
  • This code is tested on NVIDIA V100 GPU with 16GB memory
pip install -r requirements.txt

Dataset

  • Please download the PF-PASCAL, PF-WILLOW, TSS, and Internet datasets
  • Please modify the variable DATASET_DIR in config.py
  • Please modify the variable CSV_DIR in config.py

Training

  • You may determine which dataset to be the training set by changing the $DATASET variable in train.sh
  • You may change the $BATCH_SIZE variable in train.sh to a suitable value based on the GPU memory
  • The trained model will be saved under the trained_models folder
sh train.sh

Evaluation

  • You may determine which dataset to be evaluated by changing the $DATASET variable in eval.sh
  • You may change the $BATCH_SIZE variable in eval.sh to a suitable value based on the GPU memory
sh eval.sh

Acknowledgement

You can’t perform that action at this time.