An unofficial implementation of non-local deep features for salient object detection
An unofficial implementation of Non-Local Deep Features for Salient Object Detection.
The official Tensorflow version: NLDF
Some thing difference:
The information of Loss:
Performance:
Dataset | max F(paper) | MAE(paper) | max F(here) | MAE(here) |
---|---|---|---|---|
MSRA-B | 0.911 | 0.048 | 0.9006 | 0.0592 |
Note:
git clone [email protected]:AceCoooool/NLDF-pytorch.git
cd NLDF-pytorch/
Note: the original paper use other datasets.
Download the ECSSD dataset.
bash download.sh
cd tools/
python extract_vgg.py
cd ..
python demo.py --demo_img='your_picture' --trained_model='pre_trained pth' --cuda=True
Note:
weights
directory.png/demo.jpg
python main.py --mode='train' --train_path='you_data' --label_path='you_label' --batch_size=8 --visdom=True --area=True
Note:
--area=True, --boundary=True
area and boundary Dice IOU (default: --area=True --boundary=False
)--val=True
add the validation (but your need to add the --val_path
and --val_label
)you_data, you_label
means your training data root. (connect to the step 2)python main.py --mode='test', --test_path='you_data' --test_label='your_label' --batch_size=1 --model='your_trained_model'
Note:
inf
,it is better to use area Dice IOU.Maybe, it is better to add Batch Normalization.