Skip to content

Latest commit

 

History

History
47 lines (31 loc) · 1.55 KB

README.md

File metadata and controls

47 lines (31 loc) · 1.55 KB

Progressive Diversity Generation for Single Domain Generalization (PDG)

PDG extends the previous work "Diversity Probe" (ACM MM2023) by incorporating the $f$-diversity and the progressive generative framework to generate potential images from diverse perspectives.

Paper Link: https://dl.acm.org/doi/abs/10.1145/3581783.3612375

This paper appears in: IEEE Transactions on Multimedia (TMM)

Environment

  • python=3.9.16
  • torch==2.0.1
  • torchvision==0.15.2
  • munkres=1.1.4
  • numpy==1.24.1
  • opencv-python==4.7.0.72
  • scikit-learn=1.2.2
  • pandas==2.0.1

Setting up the data

Note: You need to download the data if you wish to train your own model.

Download the digits dataset from this link[BaiDuYunDisk] and its extracted code: xcl3. Please extract it inside the data directory

cd data
unzip digits.zip
cd ..

Evaluating the model

Pretrained task model is available at this link[BaiDuYunDisk] and its extracted code:2mz0. Download and extract it in the models_pth directory.

In train.py:

  • Specify the output directory to save the results in --dir.
  • Turn on the evaluation in--eval
  • Run python train.py --dir SAVE_DIR --eval True

Acknowledgement

We thank the following authors for releasing their source code, data and models: