-58- 
DETAILED 3D RECONSTRUCTION OF MONUMENTS USING MULTIPLE 
TECHNIQUES 
Sabry F. El-Hakim, J.-Angelo Beraldin, Michel Picard 
Visual Information Technology (VIT) Group 
Institute For Information Technology, National Research Council Canada (NRC) 
Ottawa, Ontario, Canada K1A 0R6 
E-mail: {Sabry.El-Hakim; Angelo.Beraldin; Michel.Picard}@nrc.ca 
Commission V, Working Group V/2 
KEY WORDS: Cultural Heritage, Laser scanning, Three-dimensional, Reconstruction, Texture, Virtual Reality, Registration 
ABSTRACT: 
The use of 3D digitization and modeling in documenting heritage sites has increased significantly over the past few years. This is 
mainly due to advances in laser scanning techniques, 3D modeling software, image-based-modeling techniques, computer power, and 
virtual reality. There are many approaches currently available. The most common remains based on surveying and CAD tools and/or 
traditional photogrammetry with control points and a human operator. This is very time consuming and can be tedious and sustained 
effort. Lately, modeling methods based on scanners data and more automated image-based technique are becoming available. We 
will discuss each approach and point out its advantages and disadvantages. We will then present our approach, which is a 
combination of several technologies. The approach presented in this paper uses both interactive and automatic techniques, each 
where it is best suited, to accurately and completely model heritage objects and sites. A highly detailed structure or site can be 
modeled at various levels of detail. Image-based modeling may be used for the basic shape and main structural elements, and laser 
scanning for fine details and sculpted surfaces. The results of applying this approach were very encouraging and several models were 
created from sites all over the world. Modeling of the Abbey of Pomposa near Ferrara, Italy, will be presented as an example. 
1. INTRODUCTION 
Many cultural heritage applications require 3D reconstruction 
of real world objects and scenes. The motives are numerous: 
1. To document historic buildings, sites, and objects for 
reconstruction or restoration if they are ever destroyed, 
for example by fire, earthquake, flood, war, or erosion. 
2. To create education resources for history and culture 
students and researchers. 
3. To reconstruct historic monuments that no longer exist, 
or partially exist. 
4. To Visualize scenes from viewpoints that are impossible 
in the real world due to size or surrounding objects. 
5. To Interact with objects without risk of damage. 
6. Virtual tourism and virtual museum. 
In general, most applications specify a number of requirements: 
1. High geometric accuracy 
2. Capturing all details 
3. Photo-realism 
4. Full automation 
5. Low cost 
6. Portability 
7. Flexibility in applications 
8. Efficiency in model size 
The order of importance of these requirements depends on the 
application, for example whether it is documentation or virtual 
museum, but in many all are important. A single system that 
satisfies all requirements is still in the future. In particular, 
accurately capturing all details with a fully automated system 
for a wide range of objects and scene remains elusive. For small 
and medium sized objects, up to the size of human or a statue, 
range-based techniques such as laser scanners can provide 
accurate and complete details with high degree of automation 
[Beraldin et al, 1999], but being relatively new technology that 
is not produced in large quantities, they remain costly. They are 
also not portable enough for a single person to carry around and 
use in a manner similar to a video or digital camera. The 
resulting model can also be inefficient for large objects. Image 
based approaches entail widely available hardware and 
potentially the same system can be used for a wide range of 
objects and scenes. They are also capable of producing realistic 
looking models and those based on photogrammetry have high 
geometric accuracy. The issues that remain are the capture of 
details on unmarked and sculpted surfaces and full automation. 
Approaches that skip the geometric modeling step, such as 
image-based rendering [Kang, 1999], are good for visualization 
and limited walkthrough. However, the lack of geometric model 
impedes the accuracy and the freedom to render the 
environment from arbitrary viewpoints 
Most documented projects on cultural heritage have used one 
method or another, whereas very few have used a combination 
of techniques. For example, a group from IBM [Bernardini et 
al, 2002] used a combination of structured light 3D sensing and 
photometric stereo to model Michelangelo’s Florentine Pietà. 
Combining laser scanning with image-based modeling and 
rendering [Sequeira et al, 2001] and image-based modeling with 
image-based rendering [Debevec et al, 1996] have also been 
reported. Our approach combines the following techniques: 
• The basic shape and large regularly shaped details, like 
columns, blocks and archways, are constructed from high- 
resolution digital images. This is based on advanced 
photogrammetry with several automated features that take 
advantage of properties found in heritage structures. 
• Fine geometric details, like sculpted and irregularly shaped 
surfaces, are obtained by laser scans. This is combined and 
integrated with the basic model created in the first step. 
• Visual details on the geometric model are obtained from 
image textures and reflectance models.