Yushu-Li / OWTTT

[ICCV 2023 Oral] Official repository for “On the Robustness of Open-World Test-Time Training: Self-Training with Dynamic Prototype Expansion”

OWTTT

This repository is an official implementation for our [ICCV 2023 Oral] paper.

On the Robustness of Open-World Test-Time Training: Self-Training with Dynamic Prototype Expansion

Yushu Li¹   Xun Xu²   Yongyi Su¹   Kui Jia^1
1South China University of Technology  
²Institute for Infocomm Research (I2R), Agency for Science, Technology and Research (A*STAR)

Overview

CIFAR10-C/CIFAR100-C

The code is released in the cifar folder.

ImageNet-C/ImageNet-R

The code is released in the imagenet folder.

Citation

If you find our work useful in your research, please consider citing:

@inproceedings{
  li2023robustness,
  title={On the Robustness of Open-World Test-Time Training: Self-Training with Dynamic Prototype Expansion},
  author={Li, Yushu and Xu, Xun and Su, Yongyi and Jia, Kui},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
  month={October},
  year={2023}
}