Toshiyuki Amano Wakayama University

This study aims to conduct “”SHITSUKAN manipulation by replacing the material properties of the projection target optically,”” which is related to the research on “”SHITSUKAN generation”” of Deep SHITSUKAN.
In previous SHITSUKAN manipulations using spatially augmented reality focused on optical attributes such as glossiness and transparency, this study attempts to optically replace the material properties of the manipulation target which generates SHITSUKAN.
However, it is easy to imagine we accurately understand the original materiality of an object by simple projection mapping.
This study believes that when viewing a projected scene, if one is aware of the projection from the projector, it is possible to separate the optical relationship and reflection properties from prior knowledge, making it possible to recognize the original SHITSUKAN, for this reason.
Based on this assumption, this study aims to achieve SHITSUKAN manipulation of materiality that is not perceived as a projection by compensating for differences in brightness from the surrounding environment, reproducing high-quality and high-definition images that do not perceive pixels, and reproducing light distribution that corresponds to viewpoint changes.