Conditional Variational Autoencoder for Learned Image Reconstruction

Abstract

Learned image reconstruction techniques using deep neural networks have recently gained popularity, and have delivered promising empirical results. However, most approaches focus on one single recovery for each observation, and thus neglect the uncertainty information. In this work, we develop a novel computational framework that approximates the posterior distribution of the unknown image at each query observation. The proposed framework is very flexible: It handles implicit noise models and priors, it incorporates the data formation process (i.e., the forward operator), and the learned reconstructive properties are transferable between different datasets. Once the network is trained using the conditional variational autoencoder loss, it provides a computationally efficient sampler for the approximate posterior distribution via feed-forward propagation, and the summarizing statistics of the generated samples are used for both point-estimation and uncertainty quantification. We illustrate the proposed framework with extensive numerical experiments on positron emission tomography (with both moderate and low count levels) showing that the framework generates high-quality samples when compared with state-of-the-art methods.

arXiv comment22 pages, preliminary version appeared as 1908.01010

References in corpus (9)

• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Expectation Propagation for approximate Bayesian inference
• Weight Uncertainty in Neural Networks
• Recurrent Inference Machines for Solving Inverse Problems
• Bayesian Deep Ensembles via the Neural Tangent Kernel
• Variational Gaussian Approximation for Poisson Data
• Expectation Propagation for Poisson Data
• Probabilistic Residual Learning for Aleatoric Uncertainty in Image Restoration
• Quantifying Sources of Uncertainty in Deep Learning-Based Image Reconstruction