PyTorch implementation of our CVPR2023 paper "OpenMix: Exploring Out-of-Distribution samples for Misclassification Detection"
We run the code with torch version: 1.10.0, python version: 3.9.7
python main_cvpr.py
The outlier data can be downloaded at https://people.eecs.berkeley.edu/~hendrycks/300K_random_images.npy
If you find this useful in your research, please consider citing:
@inproceedings{zhu2023openmix,
title={OpenMix: Exploring Outlier Samples for Misclassification Detection},
author={Zhu, Fei and Cheng, Zhen and Zhang, Xu-Yao and Liu, Cheng-Lin},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
pages={12074--12083},
year={2023}
}
A list of papers that studies out-of-distribution (OOD) detection and misclassification detection (MisD)