This is the datasets of the paper "Defect Transfer GAN: Diverse Defect Synthesis for Data Augmentation" published on BMVC2022. It consists of three types of product surface and each image is labeled with its corresponding defect type. The detailed introdution of the dataset can be found in DT_GAN_BMVC_SDI_dataset.pdf
[BMVC website] [arXiv]
Figure 1. An overview of the Surface Defect Inspection (SDI) Dataset.
If you have problem cloning the repo, please install git-lfs, and install it with
git lfs install
before cloning the dataset.
If you use this dasetset please cite
@inproceedings{Wang2023DefectTG,
title={Defect Transfer GAN: Diverse Defect Synthesis for Data Augmentation},
author={Ruyu Wang and Sabrina Hoppe and Eduardo Monari and Marco F. Huber},
booktitle={British Machine Vision Conference},
year={2022},
url={https://api.semanticscholar.org/CorpusID:256900681}
}
This project is open-sourced under the CC-BY-SA-4.0 license. See the LICENSE file for details.
Please feel free to open an issue or contact us personally if you have questions, need help, or need explanations. Write to one of the following email addresses, and maybe put one other in the cc: