QuickTime VR: an image-based approach to virtual environment navigation
SIGGRAPH '95 Proceedings of the 22nd annual conference on Computer graphics and interactive techniques
A 3-D Photo Collage System for Spatial Navigations
Revised Papers from the Second Kyoto Workshop on Digital Cities II, Computational and Sociological Approaches
Distributed vision system: a perceptual information infrastructure for robot navigation
IJCAI'97 Proceedings of the 15th international joint conference on Artifical intelligence - Volume 1
Talking digital cities: connecting heterogeneous digital cities via the universal mobile interface
Digital Cities'03 Proceedings of the Third international conference on Information Technologies for Social Capital: cross-Cultural Perspectives
| Hi-index | 0.00 |
This paper proposes a new method for building a walk-through virtual space that uses the Town Digitizing tools we have developed. The virtual space is built by smoothly interpolating a large number of omnidirectional images captured from the real world. This paper discusses how to build the virtual space and describes three of our software tools. We call them TDEditor, TAG-Editor, and TDViewer. Town Digitizing is divided into four steps: omnidirectional image acquisition, parameter estimation among omnidirectional images, registration of information, and smooth interpolation among the images. A system based on Town Digitizing allows users to explore a unique virtual space that is similar to earlier 3-D graphics system. It works better than existing methods especially for complex environments that include natural objects like trees. For adding tag information to the walk-through virtual space, we introduce a new method that sets rectangular areas with information, written in HTML language, in the panoramic view. The user is presented with information associated with the object of interest on his web browser. We believe that the techniques and approach described here represent a powerful modeling method with which to build virtual cities.