| | | |

3D Technologies

Texture Mapping from 3D Models

Texture Mapping of 3D Models is part of NRC-IIT’s ongoing research in developing and applying imaging techniques for virtual representations in order to create extremely realistic visual experiences. The project Texture Mapping of 3D Models addresses the main research issues in the real-time visualization and interactive manipulation of highly textured surfaces. In this project, texture mapping is applied to two conceptions of virtual representations: photo-realistic walk-through movies, and 3D models that allow the user interactive visualization and manipulation.

Field of Application: Cultural Heritage Documentation

The texture mapping tools and techniques developed over the course of the project’s research have been applied in international collaborative research ventures in the field of cultural heritage documentation. Wall murals of frescoes in a Byzantine crypt and in a medieval tower have been the first case studies of the project. The applied techniques capture an ensured level of faithfulness to the original art to give a vista to scientists, conservationists, historians, and visitors for observing and studying important frescoed surfaces in virtual reality.

Torre Aquila in Trento's Buonconsiglio Castle

Further research has been done in collaboration with the Personal Experience with Active Cultural Heritage (PEACH) project in Trento, Italy. The aim of the international PEACH project is to create new technologies, both hardware and software, for the active appreciation of cultural heritage. An immediate goal is to generate a textured 3D model for interactive manipulation and creation of a photo-realistic walk-through movie.

NRC-IIT’s contribution to PEACH is in the area of multi-sensorial analysis and modeling of physical spaces to achieve accurate virtual reconstructions. The space being reconstructed is a tower room in Buonconsiglio Castle in Trento, whose walls are frescoed with the series of medieval paintings called the “Cycle of Months”. The frescoes depict medieval time month by month with stunning details of landscape, costume, and every element of daily life.

Research Approach

The image-based modeling approach applied in Trento used laser scanning combined with high-resolution texture maps. Due to the highly detailed contents of the frescoes, it was crucial to preserve every line or edge and all the minute painting details. Since photo-realism is of utmost importance, it became apparent that many problems related to geometric and radiometric distortions had to be solved.

In preserving every line or edge and all minute details, even slightly curved surfaces pose problems of distortions and discontinuity. Line and edge textures on the models can be visibly affected by any small geometric errors. Sources of these errors can be due to the mapping between triangle plane and image plane, and improper camera calibration and image registration.

Traditional texture mapping techniques, which are simply draping static imagery over geometry, are not sufficient to represent reality. As well, most existing techniques for real-time rendering focus on the geometry rather than the texture. In image-based modeling it is not the geometry that affects real-time rendering but the texture size when using high resolution images. Since this project focuses on high quality textures, the texture size can end up being quite large, with memory limitations becoming a major issue.

Application Tools

Whenever possible, the project uses existing software and hardware to accomplish research objectives. Problems that have no ready solution are tackled with new algorithms and specially developed tools. Tools used in this project previously developed by the Visual Information Technology group of NRC-IIT include:

TexCapture, a texture mapping software which is currently licensed for distribution with Optech Inc. laser scanners
ShapeCapture, a software package for accurate 3D measurement and modeling, available from ShapeQuest
Virtual Theatre, a special software for interactive visualization, using scene graph tools.

Related NRC-IIT Publications

Research Contact

Dr. Sabry El-Hakim
Research Officer
Visual Information Technology
NRC Institute for Information Technology
1200 Montreal Road
Building M-50, Room 358
Ottawa, ON K1A 0R6
Telephone: +1 (613) 991-6381
Fax: +1 (613) 952-0215
E-mail: Sabry El-Hakim

Business Contact

Charles Gauthier
Business Development Officer
Business Development Office, NCR
NRC Institute for Information Technology
1200 Montreal Road
Ottawa, ON K1A 0R6
Telephone: +1 (613) 993-2491
Fax: +1 (613) 952-7998
E-mail: Charles Gauthier