Lumigrafy: Method for Creation of Enhanced Images by Compositing Multiple-Illumination Photo-Sets (14257N)
This technology falls in the following categories of Singapore's IDA Infocomm Technology Roadmap 2012:
Lumigrafy creates bold, enhanced images from multiple illumination photos of a static scene. The technique composites parts of illuminations such as key lights, back-lights and grazing illumination to automatically create beautiful images. Boosting contrast and detail through the interplay of shading, reflections and textures, it produces enhanced compositions similar to those of a professional artist.
Working with tens to hundreds of illuminations, the technique chooses the best lighting to reveal compelling contrasts. Lumigrafy is based on deep connections with the aesthetics of visual perception. It considers the perception of shading, reflections and color features such as textures in the depiction of a scene. The technique finds and assembles parts of the lighting similar to how a professional artist would, however all done algorithmically. It produces images showing well-illuminated, emphasized details and contrasts to reveal subject features.
User studies confirm Lumigrafy provides an improvement over existing techniques. Moreover, participants prefer Lumigrafy over artist compositions in more than 80% of studied cases. The surveys follow a thorough methodology and involve a substantial number of participants (219) from around the world.
For more information on this technology contact:
Technology Readiness Level 4 on the scale by the Ministry of Defence Singapore.
About the Research Group
Dr. Terence Sim is an Associate Professor at the School of Computing, NUS. His main research interests are in Face Image Analysis, Biometrics and Computational Photography.
Vlad Hosu received a PhD from NUS. Together with Prof. Sim they developed computational techniques to augment the perception of images. This has led to the creation of Lumigrafy (lumigrafy.com).