Cross-modal Hierarchical Interaction Network for RGB-D Salient Object Detection

Figure.1 The overall architecture of the proposed HINet model.
The paper can be downloaded from here[code:NEPU], which is published in Patern Recognition 🎆.

1.Requirements

Python v3.6, Pytorch 0.4.0+, Cuda 10.0, TensorboardX 2.0, opencv-python

2.Data Preparation

Download the test data from here[code:NEPU], test_in_train data from here[code:NEPU]. Then put them under the following directory:

-Dataset\   
   -train\  
   -test\ 
       -NLPR\
       -STERE\
       -SSD\
       -LFSD\
       -NJUD\
   -test_in_train\

3.Training/Testing & Evaluating

Training the HINet

Please download the released code and then:

run python Train.py

Testing the HINet

Please complete the training or download the pre-trained weights from here[code:NEPU] or Google, and then:

run python Test.py

Then the test maps will be saved to './Salmaps/'

Evaluate the result maps

You can evaluate the result maps using the tool from here[code:NEPU], thanks for Dengpin Fan.

4.Results

Qualitative comparison

Figure.2 Qualitative comparison of our proposed method with some SOTA methods.

Quantitative comparison

Table.1 Quantitative comparison with some SOTA models on five public RGB-D benchmark datasets.

Salmaps
The salmaps of the above datasets can be download from here [code:NEPU] or Google.

5.Citation

Thank you for your interest in our work, please cite:

@article{BI2022109194, title = {Cross-modal Hierarchical Interaction Network for RGB-D Salient Object Detection}, journal = {Pattern Recognition}, pages = {109194}, year = {2022}, issn = {0031-3203}, doi = {https://doi.org/10.1016/j.patcog.2022.109194}, url = {https://www.sciencedirect.com/science/article/pii/S0031320322006732}, }

6.Contact

If you have any questions, feel free to contact us via tianzhu.xiang19@gmail.com (T.-Z. Xiang) or wuranwan2020@sina.com (Ranwan Wu). For more related work, you can also visit tianzhu.xiang