Code release for FCNet: Extracting Undistorted Images for Fine-Grained Image Classification.
python 3.8
PyTorch >= 1.3.1
torchvision >= 0.4.2
- Download datatsets for FGVC (e.g. CUB-200-2011, Standford Cars, FGVC-Aircraft, etc) and organize the structure as follows:
dataset
├── train
│ ├── class_001
| | ├── 1.jpg
| | ├── 2.jpg
| | └── ...
│ ├── class_002
| | ├── 1.jpg
| | ├── 2.jpg
| | └── ...
│ └── ...
└── test
├── class_001
| ├── 1.jpg
| ├── 2.jpg
| └── ...
├── class_002
| ├── 1.jpg
| ├── 2.jpg
| └── ...
└── ...
-
Train the Stage-1 of FCNet with
python Birds.py
inFCNet_Training
-
Train the Stage-2 of FCNet with
python Birds.py
inFCNet_Training
by setting--train_stage
todefault=2
-
Train ResNet-50 integrated with FCNet with
python Birds.py
inFT_ResNet_FCNet
Thanks for your attention! If you have any suggestion or question, you can leave a message here or contact us directly: