/OOD

CVPR2021: Out-of-Distribution Detection Using Union of 1-Dimensional Subspaces

Primary LanguagePython

Out-of-Distribution Detection Using Union of 1-Dimensional Subspaces

Implementation of the out-of-distribution detection method proposed in:

A. Zaeemzadeh, N. Bisagno, Z. Sambugaro, N. Conci, N. Rahnavard, and M. Shah: Out-of-Distribution Detection Using Union of 1-Dimensional Subspaces. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021. link

  • For experiments on UCF101 video dataset see here.

Running the code

Tested on:

  • Python 3.9
  • cuda 11.2
  • torch 1.8.1
  • torchvision 0.9.1
  • numpy 1.20.1
  • sklearn 0.24.1

Downloading Out-of-Distribtion Datasets

See this repo.

Sample Script

See main.sh.

Citing this work

If you use this work in your research, please use the following BibTeX entry.

@inproceedings{zaeemzadeh2021ood,
    title={Out-of-Distribution Detection Using Union of 1-Dimensional Subspaces},
    year = {2021},
    booktitle = {Computer Vision and Pattern Recognition, 2021. CVPR 2021. IEEE Conference on},
    author={Zaeemzadeh, Alireza and Bisagno, Niccol{\`o} and Sambugaro, Zeno and Conci, Nicola and Rahnavard, Nazanin and Shah, Mubarak}
}