---
license: mit
tags:
  - anomaly-detection
  - medical-imaging
  - pytorch
  - autoencoder
  - q-former
---

# Q-Former Autoencoder (QFAE) - Pretrained Checkpoints

Official pretrained checkpoints for **"Q-Former Autoencoder: A Modern Framework for Medical Anomaly Detection"**

[![arXiv](https://img.shields.io/badge/arXiv-2507.18481-b31b1b.svg)](https://arxiv.org/abs/2507.18481)
[![GitHub](https://img.shields.io/badge/GitHub-Code-blue.svg)](https://github.com/emirhanbayar/QFAE)

## Checkpoints

| Dataset | Filename |
|---------|----------|
| Brain MRI | `brain/last_ckpt.pth` |
| Liver CT | `liver/last_ckpt.pth` |
| Chest X-Ray | `chest/last_ckpt.pth` |
| RESC | `resc/last_ckpt.pth` |


See the [GitHub repo](https://github.com/emirhanbayar/QFAE) for full inference code.

## Citation
```bibtex
@misc{dalmonte2025qformerautoencodermodernframework,
      title={Q-Former Autoencoder: A Modern Framework for Medical Anomaly Detection}, 
      author={Francesco Dalmonte and Emirhan Bayar and Emre Akbas and Mariana-Iuliana Georgescu},
      year={2025},
      eprint={2507.18481},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2507.18481}, 
}
```