Learning to Deblur Images with Exemplars

Jinshan Pan, Wenqi Ren, Zhe Hu, Ming Hsuan Yang

Research output: Contribution to journalArticle

3 Citations (Scopus)

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 algorithm for deblurring face images. In addition, we show that the proposed algorithms can be applied to image deblurring for other object classes.

Original languageEnglish
Article number8353166
Pages (from-to)1412-1425
Number of pages14
JournalIEEE transactions on pattern analysis and machine intelligence
Volume41
Issue number6
DOIs
Publication statusPublished - 2019 Jun 1

Fingerprint

Deblurring
Face
Image Deblurring
Kernel Estimation
Restoration
Neural networks
Learning
Heuristics
Neural Networks
Experiments
Demonstrate
Experiment

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics

Cite this

Pan, Jinshan ; Ren, Wenqi ; Hu, Zhe ; Yang, Ming Hsuan. / Learning to Deblur Images with Exemplars. In: IEEE transactions on pattern analysis and machine intelligence. 2019 ; Vol. 41, No. 6. pp. 1412-1425.
@article{7d9f6f4ad5d54c35af620a23c24b2fee,
title = "Learning to Deblur Images with Exemplars",
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 algorithm for deblurring face images. In addition, we show that the proposed algorithms can be applied to image deblurring for other object classes.",
author = "Jinshan Pan and Wenqi Ren and Zhe Hu and Yang, {Ming Hsuan}",
year = "2019",
month = "6",
day = "1",
doi = "10.1109/TPAMI.2018.2832125",
language = "English",
volume = "41",
pages = "1412--1425",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "6",

}

Learning to Deblur Images with Exemplars. / Pan, Jinshan; Ren, Wenqi; Hu, Zhe; Yang, Ming Hsuan.

In: IEEE transactions on pattern analysis and machine intelligence, Vol. 41, No. 6, 8353166, 01.06.2019, p. 1412-1425.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Learning to Deblur Images with Exemplars

AU - Pan, Jinshan

AU - Ren, Wenqi

AU - Hu, Zhe

AU - Yang, Ming Hsuan

PY - 2019/6/1

Y1 - 2019/6/1

N2 - 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 algorithm for deblurring face images. In addition, we show that the proposed algorithms can be applied to image deblurring for other object classes.

AB - 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 algorithm for deblurring face images. In addition, we show that the proposed algorithms can be applied to image deblurring for other object classes.

UR - http://www.scopus.com/inward/record.url?scp=85046345625&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046345625&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2018.2832125

DO - 10.1109/TPAMI.2018.2832125

M3 - Article

C2 - 29994046

AN - SCOPUS:85046345625

VL - 41

SP - 1412

EP - 1425

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 6

M1 - 8353166

ER -