External Patch-Based Image Restoration Using Importance Sampling

TL;DR

This paper presents a flexible patch-based image restoration method that leverages external datasets and importance sampling to efficiently approximate MMSE estimates, improving restoration quality across various noise conditions.

Contribution

It introduces a novel importance sampling approach for external patch-based image restoration, generalizing non-local means with enhanced flexibility and efficiency.

Findings

Effective in diverse noise scenarios

Works with large-scale and class-specific datasets

Improves restoration accuracy

Abstract

This paper introduces a new approach to patch-based image restoration based on external datasets and importance sampling. The Minimum Mean Squared Error (MMSE) estimate of the image patches, the computation of which requires solving a multidimensional (typically intractable) integral, is approximated using samples from an external dataset. The new method, which can be interpreted as a generalization of the external non-local means (NLM), uses self-normalized importance sampling to efficiently approximate the MMSE estimates. The use of self-normalized importance sampling endows the proposed method with great flexibility, namely regarding the statistical properties of the measurement noise. The effectiveness of the proposed method is shown in a series of experiments using both generic large-scale and class-specific external datasets.