/TCSVT24-DSCL

Primary LanguagePython

DSCL

Dual-Space Collaborative Learning

1. Introduction

This is the source code of IEEE TCSVT 2024 paper "Learning with Noisy Labels by Semantic and Feature Space Collaboration".

Comparison between the existing methods and the proposed method.:

The detailed learning framework of DSCL:

2. Requirements

  • python 3.7.16
  • pytorch 1.9.1
  • torchvision 0.10.1
  • numpy
  • scipy
  • tqdm
  • pillow
  • einops
  • ftfy
  • regex
  • ...

Citation:

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

H. Lin, Y. Li, Z. Zhang, L. Zhu, Y. Xu, Learning with Noisy Labels by Semantic and Feature Space Collaboration, IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), DOI: 10.1109/TCSVT.2024.3371513, 2024.

@inproceedings{lin2024learning,
  title={Learning with Noisy Labels by Semantic and Feature Space Collaboration},
  author={Lin, Han and Li, Yingjian and Zhang, Zheng and Zhu, Lei and Xu, Yong},
  journal={IEEE Transactions on Circuits and Systems for Video Technology},
  year={2024},
  publisher={IEEE}
}