1. title: WAFP-Net: Weighted Attention Fusion Based Progressive Residual Learning for Depth Map Super-Resolution

2. Due to the commerical request, the code of PaddlePaddle version is released, which can be found in the following link:

https://github.com/PaddlePaddle/PaddleDepth/tree/develop/Depth_super_resolution

3. @article{song2021wafp,
  title={WAFP-Net: Weighted Attention Fusion Based Progressive Residual Learning for Depth Map Super-Resolution},
  author={Song, Xibin and Zhou, Dingfu and Li, Wei and Dai, Yuchao and Liu, Liu and Li, Hongdong and Yang, Ruigang and Zhang, Liangjun},
  journal={IEEE Transactions on Multimedia},
  volume={24},
  pages={4113--4127},
  year={2021},
  publisher={IEEE}
}