File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Data-Driven Photo Editing and Enhancement

TitleData-Driven Photo Editing and Enhancement
Authors
Issue Date2013
PublisherIEEE Computer Society, Conference Publishing Services (CPS).
Citation
The 10th International Conference on Computer Graphics, Imaging and Visualization (CGIV), Macau, China, 6-8 August 2013. In Proceedings of the 10th Proceedings of the 10th International Conference on Computer Graphics, Imaging and Visualization (CGIV 2013), p. xviii How to Cite?
AbstractMany hard problems become tractable with the availability of large datasets. This is also true for digital photo editing and enhancement. In this talk, I present two pieces of work in this category. The first one is on data-driven image color theme editing, and the second one on example-based color and tone enhancement based on learned styles. It is often important for designers and photographers to convey or enhance desired color themes in their work. I present a data-driven method for enhancing a desired color theme in a digital photo. We formulate our goal as a unified optimization that simultaneously considers a desired color theme, texture-color relationships as well as automatic or user-specified color constraints. Quantifying the difference between an image and a color theme is made possible by color emotion spaces. We incorporate prior knowledge, such as texture-color relationships, extracted from a database of photographs to maintain a natural look of the edited photos. Experiments and a user study have confirmed the effectiveness of our method. Color and tone adjustments are among the most frequent image enhancement operations. In the second piece of work, our goal was to learn implicit color and tone adjustment rules from examples. We define tone and color adjustment rules as mappings, and propose to approximate complicated spatially varying nonlinear mappings in a piecewise manner. Parameters within such low-order models are trained using example images. We successfully applied our framework in two scenarios, low-quality photo enhancement by transferring the style of a high-end camera, and photo enhancement using styles learned from photographers and designers.
DescriptionKeynote Lectures
Persistent Identifierhttp://hdl.handle.net/10722/226177

 

DC FieldValueLanguage
dc.contributor.authorYu, Y-
dc.date.accessioned2016-06-14T04:25:12Z-
dc.date.available2016-06-14T04:25:12Z-
dc.date.issued2013-
dc.identifier.citationThe 10th International Conference on Computer Graphics, Imaging and Visualization (CGIV), Macau, China, 6-8 August 2013. In Proceedings of the 10th Proceedings of the 10th International Conference on Computer Graphics, Imaging and Visualization (CGIV 2013), p. xviii-
dc.identifier.urihttp://hdl.handle.net/10722/226177-
dc.descriptionKeynote Lectures-
dc.description.abstractMany hard problems become tractable with the availability of large datasets. This is also true for digital photo editing and enhancement. In this talk, I present two pieces of work in this category. The first one is on data-driven image color theme editing, and the second one on example-based color and tone enhancement based on learned styles. It is often important for designers and photographers to convey or enhance desired color themes in their work. I present a data-driven method for enhancing a desired color theme in a digital photo. We formulate our goal as a unified optimization that simultaneously considers a desired color theme, texture-color relationships as well as automatic or user-specified color constraints. Quantifying the difference between an image and a color theme is made possible by color emotion spaces. We incorporate prior knowledge, such as texture-color relationships, extracted from a database of photographs to maintain a natural look of the edited photos. Experiments and a user study have confirmed the effectiveness of our method. Color and tone adjustments are among the most frequent image enhancement operations. In the second piece of work, our goal was to learn implicit color and tone adjustment rules from examples. We define tone and color adjustment rules as mappings, and propose to approximate complicated spatially varying nonlinear mappings in a piecewise manner. Parameters within such low-order models are trained using example images. We successfully applied our framework in two scenarios, low-quality photo enhancement by transferring the style of a high-end camera, and photo enhancement using styles learned from photographers and designers.-
dc.languageeng-
dc.publisherIEEE Computer Society, Conference Publishing Services (CPS). -
dc.relation.ispartofProceedings of the 10th International Conference on Computer Graphics, Imaging and Visualization (CGIV 2013)-
dc.titleData-Driven Photo Editing and Enhancement-
dc.typeConference_Paper-
dc.identifier.emailYu, Y: yzyu@cs.hku.hk-
dc.identifier.authorityYu, Y=rp01415-
dc.identifier.doi10.1109/CGIV.2013.9-
dc.identifier.hkuros220979-
dc.identifier.spagexviii-
dc.identifier.epagexviii-
dc.publisher.placeLos Alamitos, CA, USA-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats