• Complex
  • Title
  • Keyword
  • Abstract
  • Scholars
  • Journal
  • ISSN
  • Conference
成果搜索

author:

Niu, Yuzhen (Niu, Yuzhen.) [1] (Scholars:牛玉贞) | Wu, Jianbin (Wu, Jianbin.) [2] | Liu, Wenxi (Liu, Wenxi.) [3] (Scholars:刘文犀) | Guo, Wenzhong (Guo, Wenzhong.) [4] (Scholars:郭文忠) | Lau, Rynson W. H. (Lau, Rynson W. H..) [5]

Indexed by:

EI SCIE

Abstract:

Synthesizing high dynamic range (HDR) images from multiple low-dynamic range (LDR) exposures in dynamic scenes is challenging. There are two major problems caused by the large motions of foreground objects. One is the severe misalignment among the LDR images. The other is the missing content due to the over-/under-saturated regions caused by the moving objects, which may not be easily compensated for by the multiple LDR exposures. Thus, it requires the HDR generation model to be able to properly fuse the LDR images and restore the missing details without introducing artifacts. To address these two problems, we propose in this paper a novel GAN-based model, HDR-GAN, for synthesizing HDR images from multi-exposed LDR images. To our best knowledge, this work is the first GAN-based approach for fusing multi-exposed LDR images for HDR reconstruction. By incorporating adversarial learning, our method is able to produce faithful information in the regions with missing content. In addition, we also propose a novel generator network, with a reference-based residual merging block for aligning large object motions in the feature domain, and a deep HDR supervision scheme for eliminating artifacts of the reconstructed HDR images. Experimental results demonstrate that our model achieves state-of-the-art reconstruction performance over the prior HDR methods on diverse scenes.

Keyword:

Dynamic range Fuses generative adversarial networks Generators High dynamic range imaging Image reconstruction Image restoration Imaging Merging multi-exposed imaging

Community:

  • [ 1 ] [Niu, Yuzhen]Fuzhou Univ, Coll Math & Comp Sci, Fujian Key Lab Network Comp & Intelligent Informa, Fuzhou 350108, Fujian, Peoples R China
  • [ 2 ] [Niu, Yuzhen]Minist Educ, Key Lab Spatial Data Min & Informat Sharing, Fuzhou 350108, Fujian, Peoples R China
  • [ 3 ] [Wu, Jianbin]Fuzhou Univ, Coll Math & Comp Sci, Fuzhou 350000, Peoples R China
  • [ 4 ] [Liu, Wenxi]Fuzhou Univ, Coll Math & Comp Sci, Fuzhou 350000, Peoples R China
  • [ 5 ] [Guo, Wenzhong]Fuzhou Univ, Coll Math & Comp Sci, Fuzhou 350000, Peoples R China
  • [ 6 ] [Lau, Rynson W. H.]City Univ Hong Kong, Dept Comp Sci, Hong Kong, Peoples R China

Reprint 's Address:

  • 刘文犀

    [Liu, Wenxi]Fuzhou Univ, Coll Math & Comp Sci, Fuzhou 350000, Peoples R China

Show more details

Version:

Related Keywords:

Related Article:

Source :

IEEE TRANSACTIONS ON IMAGE PROCESSING

ISSN: 1057-7149

Year: 2021

Volume: 30

Page: 3885-3896

1 1 . 0 4 1

JCR@2021

1 0 . 8 0 0

JCR@2023

ESI Discipline: ENGINEERING;

ESI HC Threshold:105

JCR Journal Grade:1

CAS Journal Grade:1

Cited Count:

WoS CC Cited Count: 113

SCOPUS Cited Count: 150

ESI Highly Cited Papers on the List: 0 Unfold All

WanFang Cited Count:

Chinese Cited Count:

30 Days PV: 2

Online/Total:69/10011705
Address:FZU Library(No.2 Xuyuan Road, Fuzhou, Fujian, PRC Post Code:350116) Contact Us:0591-22865326
Copyright:FZU Library Technical Support:Beijing Aegean Software Co., Ltd. 闽ICP备05005463号-1