- Code release for 《Multi-Frequency Feature Enhancement for Multi-Granularity Visual Classification》
- python 3.6
- PyTorch 1.2.0
- torchvision
- Download datasets
- Extract them to data/Birds/, data/Airs/ and data/Cars/, respectively.
- Split the dataset into train and test folder, the index of each class should follow the /MGVC dataset list
- Just run main.py
- The parameters you see are training CUB-200-2011 dataset in main.py. If you want to run Stanford Cars and FGVC-Aircraft datasets, you can adjust them directly in the main.py
Thanks for your attention! If you have any suggestion or question, you can leave a message here or contact us directly: