Home

Awesome

<div align="center"> <h1> AutoView </h1> <span><font size="5", > Learning Self-Regularized Adversarial Views for Self-Supervised Vision Transformers </font></span> </br> Tao Tang∗, Changlin Li∗, Guangrun Wang, Kaicheng Yu, Xiaojun Chang, Xiaodan Liang<sup><span>&#8224;</span></sup>

(<span>*</span>: equal contribution, <span></span>: corresponding author)

<br> <div><a href="https://arxiv.org/pdf/2210.08458.pdf">[arXiv Preprint]</a></div> </div>

Introduction

Framework

We propose AutoView, a self-regularized adversarial AutoAugment method, to learn views for self-supervised vision transformers.

Visualization

vis

Getting Started

git clone https://github.com/Trent-tangtao/AutoView.git

This is a preliminary release. We have not carefully organized everything now.

Citation

If you find AutoView is useful in your research or applications, please consider giving us a star 🌟 and citing it by the following BibTeX entry.

@article{tang2022learning,
  title={Learning Self-Regularized Adversarial Views for Self-Supervised Vision Transformers}, 
  author={Tao Tang and Changlin Li and Guangrun Wang and Kaicheng Yu and Xiaojun Chang and Xiaodan Liang},
  journal={arXiv preprint arXiv:2210.08458},
  year={2022}
}