Learning to Deblur Images with Exemplars

TL;DR

This paper introduces a novel face image deblurring method that leverages facial structures and a neural network to restore sharp edges, outperforming existing techniques and applicable to other object classes.

Contribution

The paper presents a new deblurring algorithm using an exemplar dataset and a CNN for edge restoration, avoiding traditional coarse-to-fine strategies.

Findings

Outperforms state-of-the-art face deblurring methods

Effective in restoring facial details from blurry images

Applicable to deblurring other object classes

Abstract

Human faces are one interesting object class with numerous applications. While significant progress has been made in the generic deblurring problem, existing methods are less effective for blurry face images. The success of the state-of-the-art image deblurring algorithms stems mainly from implicit or explicit restoration of salient edges for kernel estimation. However, existing methods are less effective as only few edges can be restored from blurry face images for kernel estimation. In this paper, we address the problem of deblurring face images by exploiting facial structures. We propose a deblurring algorithm based on an exemplar dataset without using coarse-to-fine strategies or heuristic edge selections. In addition, we develop a convolutional neural network to restore sharp edges from blurry images for deblurring. Extensive experiments against the state-of-the-art methods demonstrate the effectiveness of the proposed algorithms for deblurring face images. In addition, we show the proposed algorithms can be applied to image deblurring for other object classes.