Neural Maximum A Posteriori Estimation on Unpaired Data for Motion Deblurring

TL;DR

This paper introduces NeurMAP, a novel neural maximum a posteriori framework that trains deblurring networks on unpaired data by jointly estimating motion and sharp content, overcoming limitations of prior methods.

Contribution

NeurMAP is the first framework enabling training of image deblurring networks on unpaired datasets using a joint motion estimation and deblurring approach.

Findings

Outperforms state-of-the-art methods in quantitative metrics

Achieves superior visual quality in deblurring results

Effectively models motion and sharp content from unpaired data

Abstract

Real-world dynamic scene deblurring has long been a challenging task since paired blurry-sharp training data is unavailable. Conventional Maximum A Posteriori estimation and deep learning-based deblurring methods are restricted by handcrafted priors and synthetic blurry-sharp training pairs respectively, thereby failing to generalize to real dynamic blurriness. To this end, we propose a Neural Maximum A Posteriori (NeurMAP) estimation framework for training neural networks to recover blind motion information and sharp content from unpaired data. The proposed NeruMAP consists of a motion estimation network and a deblurring network which are trained jointly to model the (re)blurring process (i.e. likelihood function). Meanwhile, the motion estimation network is trained to explore the motion information in images by applying implicit dynamic motion prior, and in return enforces the deblurring network training (i.e. providing sharp image prior). The proposed NeurMAP is an orthogonal approach to existing deblurring neural networks, and is the first framework that enables training image deblurring networks on unpaired datasets. Experiments demonstrate our superiority on both quantitative metrics and visual quality over state-of-the-art methods. Codes are available on https://github.com/yjzhang96/NeurMAP-deblur.