This project link: https://github.com/lartpang/CMWNet.pytorch

This project provides an unofficial pytorch implementation of the RGB-D SOD method for 'Cross-Modal Weighting Network for RGB-D Salient Object Detection', ECCV 2020.

• Paper Link: https://arxiv.org/pdf/2007.04901.pdf
• Official Caffe Code: https://github.com/MathLee/CMWNet

• torch >= 1.5.0 for AMP.
• torchvision
• numpy
• tqdm
• opencv-python
• pillow
• py_sod_metrics for evaluating the results.

1. Configure your datasets in dataset_config.py.
2. Set hyperparameters in main.py.
3. python main.py
4. All results will be saved into the ckpt_path (see main.py).