Bayesian Image Reconstruction using Deep Generative Models

8 Dec 2020  ·  Razvan V Marinescu, Daniel Moyer, Polina Golland ·

Machine learning models are commonly trained end-to-end and in a supervised setting, using paired (input, output) data. Examples include recent super-resolution methods that train on pairs of (low-resolution, high-resolution) images. However, these end-to-end approaches require re-training every time there is a distribution shift in the inputs (e.g., night images vs daylight) or relevant latent variables (e.g., camera blur or hand motion). In this work, we leverage state-of-the-art (SOTA) generative models (here StyleGAN2) for building powerful image priors, which enable application of Bayes' theorem for many downstream reconstruction tasks. Our method, Bayesian Reconstruction through Generative Models (BRGM), uses a single pre-trained generator model to solve different image restoration tasks, i.e., super-resolution and in-painting, by combining it with different forward corruption models. We keep the weights of the generator model fixed, and reconstruct the image by estimating the Bayesian maximum a-posteriori (MAP) estimate over the input latent vector that generated the reconstructed image. We further use variational inference to approximate the posterior distribution over the latent vectors, from which we sample multiple solutions. We demonstrate BRGM on three large and diverse datasets: (i) 60,000 images from the Flick Faces High Quality dataset (ii) 240,000 chest X-rays from MIMIC III and (iii) a combined collection of 5 brain MRI datasets with 7,329 scans. Across all three datasets and without any dataset-specific hyperparameter tuning, our simple approach yields performance competitive with current task-specific state-of-the-art methods on super-resolution and in-painting, while being more generalisable and without requiring any training. Our source code and pre-trained models are available online:

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Denoising FFHQ BRGM LPIPS 0.24 # 1
Image Inpainting FFHQ 1024 x 1024 BRGM LPIPS 0.19 # 1
RMSE 24.28 # 1
PSNR 21.33 # 1
SSIM 0.84 # 1
Image Inpainting FFHQ 1024 x 1024 SN-PatchGAN LPIPS 0.24 # 2
RMSE 30.75 # 2
PSNR 19.67 # 2
SSIM 0.82 # 2
Image Super-Resolution FFHQ 256 x 256 - 4x upscaling BRGM PSNR 24.16 # 3
SSIM 0.70 # 6
Image Denoising FFHQ 64x64 - 4x upscaling BRGM LPIPS 0.24 # 1