gulation is performed: 
surface model of the 
sforming the cartesian 
e frame. In the defined 
inder axis crossing the 
Y-axis of the cartesian 
lean distance from the 
1gle around the z-axis. 
OC plane and the final 
shown in Figure 9. 
  
ed model 
set is mapped onto the 
realistic virtual model 
> not modeled because 
visible. 
      
te 
the Great Buddha 
3.3 Metric images - Manual measurements 
With the manual measurement a point cloud of ca 28000 points 
is obtained. In the point visualization of Figure 11 it is already 
possible to distinguish the shapes of the folds on the dress. 
ze 
/ 
} 
í 
  
   
Figure 11: The point cloud of the manual measurement. 
The main edges and the structures of the folds, measured as 
breaklines, are well visible 
The following surface triangulation is able to reconstruct the 
features of the dress (Figure 12). The final 3-D model is 
presented in Figure 13. 
  
Figure 12: Visualization in wireframe mode of the 3-D 
structures on the central part of the dress of the Buddha 
4. PHYSICAL RECONSTRUCTION 
The 3-D computer model that we reconstructed with the manual 
procedure is used for the physical reconstruction of the Great 
Buddha. At the Institute of Virtual Manufacturing, ETH Zurich, 
R.Zanini and J. Wirth have recreated a 1:200 model statue of the 
Great Buddha. The point cloud of the photogrammetric 
reconstruction is imported in a digitally programmed machine 
tool (Starrag NF100) without any further processing (Wirth, 
2002). The machine works on polyurethane boxes and follows 
milling paths calculated directly from the point cloud. The 
  
physical model is recreated in three steps: (1) a roughing path, 
(2) a pre-smoothing path and (3) the final smoothing path. The 
time needed for preparing the production data was about 3 
hours while the milling of the part itself was done in about 8 
hours. 
  
Figure 13: The texturized 3-D model of the statue created 
with manual measurements on the metric images 
5. CONCLUSIONS 
The computer reconstruction of the Great Buddha of Bamiyan, 
Afghanistan has been performed successfully using various 
digital photogrammetric techniques. We have presented here 
three versions of the 3D model, based on (a) automated point 
cloud generation using four internet images, (b) automated 
point cloud generation using three metric images, (c) manual 
measurements using three metric images. While the automated 
matching methods provide for dense point clouds, they fail to 
model the very fine details of the statue, e.g. the folds of the 
robe. Also, some important edges are missed. Only manual 
measurements allow to generate a 3-D model which is accurate 
and complete enough to serve as the basis for the physical 
reconstruction. Therefore, we will use the results of version (c) 
for the physical reconstruction of the statue. With a pixel size of 
10 micron (1 cm on the object) manual measurements can be 
done with a relative accuracy of about 1-2 cm. While such high 
accuracy is necessary to model the folds (5 - 10 cm in size) 
correctly, it is surely more than sufficient to represent the 
overall form of the 53 m high statue in very close resemblance 
to the original. The problems encountered here with the 
orientation of amateur images and with automated matching 
could be solved in an acceptable manner. The main difficulties 
of this project consisted in the transition from the point cloud 
(including breaklines) to a surface model which can satisfy high 
modeling and visualization demands. Since automated image 
matching does not take into consideration the geometrical 
object surface conditions it is very difficult to turn such more or 
less randomly generated point clouds into TIN or wireframe 
structures of high quality and without losing essential 
information. Even when measurements are done in manual 
mode it is crucial for the operator to understand the functional 
behaviour of the subsequently activated 3-D modeler. In this 
—367—