Home

Awesome

<div align="center">

[ECCV 2024] An Incremental Unified Framework for Small Defect Inspection

This is the official repository for IUF (ECCV 2024).

Jiaqi Tang, Hao Lu, Xiaogang Xu, Ruizheng Wu, Sixing Hu,

Tong Zhang, Twz Wa Cheng, Ming Ge, Ying-Cong Chen* and Fugee Tsung.

*: Corresponding Author

GitHub license made-for-VSCode Visits Badge

Here is our Project Page with Video!

</div>

šŸ” Our Setting: Incremental Unified Framework (IUF)

<div align="center"> <img src="imgs/setting.png" alt="Our Setting"> </div>

šŸ“¢ News and Updates

ā–¶ļø Getting Started

<!-- 1. [Installation](#installation) 2. [Dataset](#dataset) 3. [Configuration](#configuration) 5. [Testing](#Testing) 4. [Training](#Training) -->

šŸŖ’ Installation

šŸ’¾ Dataset Preparation

šŸ”Ø Configuration

šŸ–„ļø Training and Testing

āš” Performance

Compared with other baselines, our model achieves state-of-the-art performance:

ā­ [Figure 1] Quantitative evaluation in MvTecAD.

ā­ [Figure 2] Quantitative evaluation in VisA.

ā­ [Figure 3] Qualitative Evaluation.

šŸŒ Citations

The following is a BibTeX reference:

@inproceedings{tang2024incremental,
  title = {An Incremental Unified Framework for Small Defect Inspection},
  author = {Tang, Jiaqi and Lu, Hao and Xu, Xiaogang and Wu, Ruizheng and Hu, Sixing and Zhang, Tong and Cheng, Tsz Wa and Ge, Ming and Chen, Ying-Cong and Tsung, Fugee},
  booktitle = {18th European Conference on Computer Vision (ECCV)},
  year = {2024}
}

šŸ“§ Connecting with Us?

If you have any questions, please feel free to send email to jtang092@connect.hkust-gz.edu.cn.

šŸ“œ Acknowledgment

The research work was sponsored by AIR@InnoHK. The code is inspired by UniAD.