/bianet

Bilateral Attention Network for RGB-D Salient Object Detection

Primary LanguagePython


Logo

Bilateral Attention Network for RGB-D Salient Object Detection

Published in IEEE Transactions on Image Processing (TIP)
[Paper 📄] [ArXiv 🌐] [Homepage 🏠] »


Prerequisites

Environments

  • PyTorch >= 1.0
  • Ubuntu 18.04

Usage

  1. Download the model parameters and datasets
  2. Configure test.sh
--backbones vgg16+vgg11+res50+res2_50 (Multiple items are connected with '+')
--datasets dataset1+dataset2+dataset3
--param_root param (pretrained model path)
--input_root your_data_root (categorize by subfolders)
--save_root your_output_root
  1. Run by
sh test.sh

Model parameters and prediction results

Model parameters Prediction results
VGG-16 [Google Drive] [Baidu Pan (bfrn)] [Google Drive] [Baidu Pan (k01w)]
VGG-11 [Google Drive] [Baidu Pan (2a5c)] [Google Drive] [Baidu Pan (d0t7)]
ResNet-50 [Google Drive] [Baidu Pan (o9l2)] [Google Drive] [Baidu Pan (dqw1)]
Res2Net-50 [Google Drive] [Baidu Pan (k761)] [Google Drive] [Baidu Pan (h3t9)]

Citation

@article{zhang2020bianet,
  title={Bilateral attention network for rgb-d salient object detection},
  author={Zhang, Zhao and Lin, Zheng and Xu, Jun and Jin, Wenda and Lu, Shao-Ping and Fan, Deng-Ping},
  journal={IEEE Transactions on Image Processing (TIP)},
  volume={30},
  pages={1949-1961},
  doi={10.1109/TIP.2021.3049959},
  year={2021},
}

Contact

If you have any questions, feel free to contact me via zzhang🥳mail😲nankai😲edu😲cn