Learn from Unpaired Data for Image Restoration: A Variational Bayes Approach

Dihan Zheng au Mathematical Sciences Center, Tsinghua University, Beijing 100084, China Xiaowen Zhang Hisilicon, Shanghai 300060, China Kaisheng Ma stitute for Interdisciplinary Information Science, Tsinghua University, Beijing 100084, China Chenglong Bao Yau Mathematical Sciences Center, Tsinghua University, Beijing 100084, China, and Yanqi Lake Beijing Institute of Mathematical Sciences and Applications, Beijing 101408, China

TBD mathscidoc:2206.43001

2022.4
Collecting paired training data is difficult in practice, but the unpaired samples broadly exist. Current approaches aim at generating synthesized training data from the unpaired samples by exploring the relationship between the corrupted and clean data. This work proposes LUD-VAE, a deep generative method to learn the joint probability density function from data sampled from marginal distributions. Our approach is based on a carefully designed probabilistic graphical model in which the clean and corrupted data domains are conditionally independent. Using variational inference, we maximize the evidence lower bound (ELBO) to estimate the joint probability density function. Furthermore, we show that the ELBO is computable without paired samples under the inference invariant assumption. This property provides the mathematical rationale of our approach in the unpaired setting. Finally, we apply our method to real-world image denoising and super-resolution tasks and train the models using the synthetic data generated by the LUD-VAE. Experimental results validate the advantages of our method over other learnable approaches.
No keywords uploaded!
[ Download ] [ 2022-06-13 16:49:57 uploaded by Baocl ] [ 269 downloads ] [ 0 comments ]
@inproceedings{dihan2022learn,
  title={Learn from Unpaired Data for Image Restoration: A Variational Bayes Approach},
  author={Dihan Zheng, Xiaowen Zhang, Kaisheng Ma, and Chenglong Bao},
  url={http://archive.ymsc.tsinghua.edu.cn/pacm_paperurl/20220613164957505921347},
  year={2022},
}
Dihan Zheng, Xiaowen Zhang, Kaisheng Ma, and Chenglong Bao. Learn from Unpaired Data for Image Restoration: A Variational Bayes Approach. 2022. http://archive.ymsc.tsinghua.edu.cn/pacm_paperurl/20220613164957505921347.
Please log in for comment!
 
 
Contact us: office-iccm@tsinghua.edu.cn | Copyright Reserved