Robust deep ensemble method for real-world image denoising
Pengju Liu,Hongzhi Zhang,Jinghui Wang,Yuzhi Wang,Dongwei Ren,Wangmeng Zuo
DOI: https://doi.org/10.1016/j.neucom.2022.09.058
IF: 6
2022-11-01
Neurocomputing
Abstract:Recently, deep learning-based image denoising methods have achieved promising performance. However, when the distribution of real-world noisy images is unknown, the denoising performance is still limited due to the domain gap between the training set and the testing set. Nonetheless, the unknown noise distribution usually can be modeled as proper combination of existing noise distributions. In this paper, we propose a simple yet effective Bayesian deep ensemble (BDE) method for real-world image denoising, where several representative deep denoisers pre-trained with various training data settings can be fused to improve robustness. The foundation of BDE is that real-world image noises are highly signal-dependent, and heterogeneous noises in a real-world noisy image can be separately handled by different denoisers. In particular, we take well-trained CBDNet, NBNet, HINet, Uformer and GMSNet into denoiser pool, and a U-Net is adopted to predict pixel-wise weighting maps to fuse these denoisers. Instead of solely learning pixel-wise weighting maps, Bayesian deep learning strategy is introduced to predict weighting uncertainty as well as weighting map, by which prediction variance can be modeled for improving robustness on real-world noisy images. Extensive experiments have shown that real-world noises can be better removed by fusing existing denoisers instead of training a big denoiser with expensive cost. Furthermore, our BDE can be extended to other image restoration tasks, i.e. image deblurring, image deraining and single image super-resolution, and also achieves significance gain on benchmark datasets.
computer science, artificial intelligence