T-SVDNet: Exploring High-Order Prototypical Correlations for Multi-Source Domain Adaptation (ICCV 2021, official Pytorch implementation)

The paper is available here: Arxiv

Abstract

Most existing domain adaptation methods focus on adaptation from only one source domain, however, in practice there are a number of relevant sources that could be leveraged to help improve performance on target domain. We propose a novel approach named T-SVDNet to address the task of Multi-source Domain Adaptation (MDA), which is featured by incorporating Tensor Singular Value Decomposition (T-SVD) into a neural network's training pipeline. Overall, high-order correlations among multiple domains and categories are fully explored so as to better bridge the domain gap. Specifically, we impose Tensor-Low-Rank (TLR) constraint on a tensor obtained by stacking up a group of prototypical similarity matrices, aiming at capturing consistent data structure across different domains. Furthermore, to avoid negative transfer brought by noisy source data, we propose a novel uncertainty-aware weighting strategy to adaptively assign weights to different source domains and samples based on the result of uncertainty estimation. Extensive experiments conducted on public benchmarks demonstrate the superiority of our model in addressing the task of MDA compared to state-of-the-art methods.

Installation

pip install -r requirements.txt

Data Preparation

Download Digits-Five, DomainNet and PACS.

Training

To train reproduce the performance, simply run:

python train.py --use_target --save_model --target $target_domain$ \
                --checkpoint_dir $save_dir$

##Citation If you like our work and use the code or models for your research, please cite our work as follows.

@inproceedings{li2021t,
  title={T-SVDNet: Exploring High-Order Prototypical Correlations for Multi-Source Domain Adaptation},
  author={Li, Ruihuang and Jia, Xu and He, Jianzhong and Chen, Shuaijun and Hu, Qinghua},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
  pages={9991--10000},
  year={2021}
}

Acknowledgments

This code is heavily borrowed from LtC-MSDA and M3SDA.