Here I provide the PyTorch implementation for CLAFA.
RTX 3090
python 3.8.8
PyTorch 1.11.0
mmcv 1.6.0
Clone this repo:
git clone https://github.com/xingronaldo/CLAFA.git
cd CLAFA
- Install dependencies
All dependencies can be installed via 'pip'.
Download data and add them to ./datasets
.
Here I provide the trained models for the SV-CD dataset Baidu Netdisk, code: CLAFA.
Put them in ./checkpoints
.
- Test on the SV-CD dataset with the MobileNetV2 backbone
python test.py --backbone mobilenetv2 --name SV_mobilenetv2 --gpu_ids 1
- Test on the SV-CD dataset with the ResNet18d backbone
python test.py --backbone resnet18d --name SV_resnet18d --gpu_ids 1
python trainval.py --gpu_ids 1
All the hyperparameters can be adjusted in option.py
.