File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: All-day thin-lens computational imaging with scene-specific learning recovery

TitleAll-day thin-lens computational imaging with scene-specific learning recovery
Authors
Issue Date2022
Citation
Applied Optics, 2022, v. 61, n. 4, p. 1097-1105 How to Cite?
AbstractModern imaging optics ensures high-quality photography at the cost of a complex optical form factor that deviates from the portability. The drastic development of image processing algorithms, especially advanced neural networks, shows great promise to use thin optics but still faces the challenges of residual artifacts and chromatic aberration. In this work, we investigate photorealistic thin-lens imaging that paves the way to actual applications by exploring several fine-tunes. Notably, to meet all-day photography demands, we develop a scene-specific generative-adversarial-network-based learning strategy and develop an integral automatic acquisition and processing pipeline. Color fringe artifacts are reduced by implementing a chromatic aberration pre-correction trick. Our method outperforms existing thin-lens imaging work with better visual perception and excels in both normal-light and low-light scenarios.
Persistent Identifierhttp://hdl.handle.net/10722/315384
ISSN
2023 Impact Factor: 1.7
2023 SCImago Journal Rankings: 0.487
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorQi, Bingyun-
dc.contributor.authorChen, Wei-
dc.contributor.authorDun, Xiong-
dc.contributor.authorHao, Xiang-
dc.contributor.authorWang, Rui-
dc.contributor.authorLiu, Xu-
dc.contributor.authorLi, Haifeng-
dc.contributor.authorPeng, Yifan-
dc.date.accessioned2022-08-05T10:18:41Z-
dc.date.available2022-08-05T10:18:41Z-
dc.date.issued2022-
dc.identifier.citationApplied Optics, 2022, v. 61, n. 4, p. 1097-1105-
dc.identifier.issn1559-128X-
dc.identifier.urihttp://hdl.handle.net/10722/315384-
dc.description.abstractModern imaging optics ensures high-quality photography at the cost of a complex optical form factor that deviates from the portability. The drastic development of image processing algorithms, especially advanced neural networks, shows great promise to use thin optics but still faces the challenges of residual artifacts and chromatic aberration. In this work, we investigate photorealistic thin-lens imaging that paves the way to actual applications by exploring several fine-tunes. Notably, to meet all-day photography demands, we develop a scene-specific generative-adversarial-network-based learning strategy and develop an integral automatic acquisition and processing pipeline. Color fringe artifacts are reduced by implementing a chromatic aberration pre-correction trick. Our method outperforms existing thin-lens imaging work with better visual perception and excels in both normal-light and low-light scenarios.-
dc.languageeng-
dc.relation.ispartofApplied Optics-
dc.titleAll-day thin-lens computational imaging with scene-specific learning recovery-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1364/AO.448155-
dc.identifier.pmid35201084-
dc.identifier.scopuseid_2-s2.0-85124310571-
dc.identifier.volume61-
dc.identifier.issue4-
dc.identifier.spage1097-
dc.identifier.epage1105-
dc.identifier.eissn2155-3165-
dc.identifier.isiWOS:000749795600032-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats