Learn from Unpaired Data for Image Restoration: A Variational Bayes Approach
This addresses the challenge of collecting paired data for image restoration, which is incremental as it builds on existing unpaired learning methods.
The paper tackles the problem of learning image restoration models without paired training data by proposing LUD-VAE, a deep generative method that estimates the joint probability density from unpaired samples, and applies it to tasks like denoising and super-resolution, showing experimental advantages over other approaches.
Collecting paired training data is difficult in practice, but the unpaired samples broadly exist. Current approaches aim at generating synthesized training data from 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, super-resolution, and low-light image enhancement tasks and train the models using the synthetic data generated by the LUD-VAE. Experimental results validate the advantages of our method over other approaches.