Complementarity-Aware-Attention-Network-for-Salient-Object-Detection

This is Pytorch implementation of our TCYB(IEEE TRANSACTIONS ON CYBERNETICS) paper.

Prerequisites

• Pytorch 1.0

Usage

1. Clone the repository

git https://github.com/silencedog/Complementarity-Aware-Attention-Network-for-Salient-Object-Detection.git

2. Download the best pre-trained models and result maps

Download the following pre-trained models BaiduYun Extracted code： wbpw into CANet folder.

Download the our prediction result maps BaiduYun Extracted code： wbpw into CANet folder.

3. Train

1. Set the main.py and CANet.py path.
2. Set the mode parameters of main.py is train.

4. Test

1. Set the datasets path in main.py (Actually dataname of main.py).

5. Touch with me

If you have some question: cxpanzefeng@163.com

Thanks to DSS-pytorch

Thanks to BDCN

Thanks to DANet