SwinMCNet: Mirror Complementary Transformer Network for RGB-thermal Salient Object Detection
Download the following datasets and unzip them into data folder
Our proposed RGBT SOD dataset VT723 that contain common challenging scenes of real world.
- Split the ground truth into skeleton map and contour map, which will be saved into
data/VT5000/skeletonanddata/VT5000/contour.
python3 utils.py- Train the model and get the pretrained model, which will be saved into
resfolder.
python3 train.py- If you just want to evaluate the performance of MCNet without training, please download the pretrained model into
resfolder. - Test the model and get the predicted saliency maps, which will be saved into
mapsfolder.
python3 test.py- Evaluate the predicted results.
cd eval
matlab
main