by Xuemiao Xu^, Jiaxing Chen^, Huaidong Zhang*, and Wing~W. Y. Ng* (^ joint 1st author, * joint corresponding author)[paper link]

This implementation is written by Jiaxing Chen at the South China University of Technology.

@article{xu2020d4net,
     title={D4Net: De-Deformation Defect Detection Network for Non-Rigid Products with Large Patterns},
     author={Xuemiao Xu, Jiaxing Chen, Huaidong Zhang, and Wing~W. Y. Ng},
     journal={Information Sciences},
     volume={547},
     pages={763--776},
     year={2021},
     publisher={Elsevier}
}

Due to the influence of COVID-19, the LFLP dataset will be released after the author returns to school. [LFLP dataset link]

You can download the trained model which is reported in our paper at Google Drive.

• Python 2.7
• PyTorch 0.4.0
• torchvision
• numpy

1. Set the path of pretrained ResNeXt model in resnext/config.py
2. Set the path of LFLP dataset in config.py
3. Run by python train.py

Hyper-parameters of training were gathered at the beginning of train.py and you can conveniently change them as you need.

1. Put the trained model in ckpt/d4net
2. Run by python infer.py

Settings of testing were gathered at the beginning of infer.py and you can conveniently change them as you need.