| Research and Development of a 3-dimensional Digital Archiving Technology and its Contribution in Applications to Overseas Content |
| Katsushi Ikeuchi |
 Katsushi Ikeuchi |
@ The recipient is known as one of the pioneers in the area referred to as Physics-based vision. In the early 80s, instead of applying random filters to images in an ad-hoc manner, as was done by most vision researchers, he proposed modeling the image formation process based on rigorous physical and optical theories and solving vision problems using the inverse models obtained. One of the successful achievements is his shape-from-shading algorithm
D@ In the 90s, he extended the paradigm of Physics-based vision to object modeling. In this, he modeled the inverse process of image formation from images of the original 3D object, and using this inverse process, he set up algorithms to represent 3D objects. In the mid 90s, he and his colleagues completed algorithm sets that can be successfully applied to indoor objects. This area has grown into modeling-from-reality or image-based modeling
D@In the first decade of this century, the recipient decided to apply this modeling-from-reality method to objects which it is important to model, i.e. cultural heritage, so increasing the social impact of the modeling-from-reality method. Since many objects of cultural heritage exist outdoors, the research goal was set to model large scale objects of cultural heritage in an outdoor environment, and the recipient modeled the great Buddha of Kamakura as an example of a large outdoor cultural object. Later, the recipient expanded the collection of modeled heritage objects to include the great Buddha of Nara and the Bayon temple in Cambodia
D@Following this line of research, the recipient has encountered many challenges, and has overcome these challenges through work resulting in more than ten PhD theses produced by researchers of his team. To name a few, since large scale cultural heritage objects often require observation from very high viewpoints, a balloon sensor, which can be readily used in any place, has been invented to satisfy this requirement. A single observation only covers a part of an object of cultural heritage, so many observations are necessary. The recipient and colleagues invented a jigsaw puzzle-like program to combine all such partial observations into a complete integrated dataset in a rapid manner. During the jigsaw process, errors gradually accumulate. In order to avoid such error accumulation, it is necessary to load all the partial data into the computer. But, this can causes a memory overflow as well as a very long computational time. The recipient and a colleague avoided this by introducing parallel computation and efficient data distribution over the computer network. The appearance of an object of cultural heritage varies due to the variation of outdoor illumination. The recipient and one of his students solved this issue by considering the so-called blackbody radiation, and have come up with an algorithm to rectify the appearance variations.
Fig.1 3D Model of Bayon temple in Cambodia.
@This series of achievements by the recipient has been highly regarded in the computer vision community and he has received several prestigious awards including IEEE Fellow in 1998 and the IEEE-CS ICCV Significant Researcher Award in 2011. He was also awarded the Medal of Honor with purple ribbon by the emperor of Japan. Thus, the committee holds the recipientfs achievements in high regard and considered that these should be recognized by this award.
(Q)@T. Kanade and K. Ikeuchi, gIntroduction to the special issue on physical modeling in computer vision,h IEEE Trans. Pattern Anal. Mach. Intell., vol.13. no.7, pp.609-610, 1991.
(R)@S.K. Nayar, K. Ikeuchi, and T. Kanade, gSurface reflection:Physical and geometrical perspective,h IEEE Trans. Pattern Anal. Mach. Intell., vol.13, no.7, pp.661-634, 1991.
(S)@Y. Sato, M.D. Wheeler, and K. Ikeuchi, gObject shape and reflectance modeling from observation,h Proceedings of ACM SIGGRAPH, pp.379-387, 1997.
(T)@M.D. Wheeler, Y. Sato, and K. Ikeuchi, gConsensus surfaces for modeling 3D objects from multiple range images,h Proc in International Conference on Computer Vision, pp.917-924, 1998.
(U)@K. Ikeuchi and Y. Sato, Modeling-from-reality, Kluwer Academic, 2001.
(V)@A. Banno, T. Masuda, T. Oishi, and K. Ikeuchi, gFlying laser range sensor for large-scale site-modeling and its applications in Bayon digital archival project,h Int. J. Comput. Vis., vol.78, no.2-3, pp.207-222, 2008.
(W)@K. Ikeuchi, T. Oishi, J. Takamatsu, R. Sagawa, A. Nakazawa, R. Kurazume, K. Nishino, M. Kamakura, and Y. Okamoto, gThe great Buddha project:Digitally archiving, restoring, and analyzing cultural heritage,h Int. J. Comput. Vis., vol.75, no.1, pp.189-208, 2007.
(X)@K. Ikeuchi and D. Miyazaki, Digitally Archiving Cultural Objects, Springer, 2008.
(10)@D. Miyazaki, K. Hara, and K. Ikeuchi, gMedian photometric stereo as applied to the Segonko Tumulus and museum objects,h Int. J. Comput. Vis., vol.86, no.2-3, pp.229-242, 2010.
