of Detail, 
principals 
an elegant 
resources: 
speed eye 
& Geisler, 
of interest 
sent over a 
:hnique of 
ave been 
ents have 
aphics and 
licroscopic 
sive sensor 
sualization 
st. 
intation of 
ind terrain 
| purposes 
e. 
with great 
being fully 
ble is not 
A more 
vailable on 
nes with a 
onstructing 
; alone can 
nd photo- 
he “virtual 
igger. 
ga 
  
  
Managing such a massive dataset is difficult even with the 
games that come in stand-alone CDs, because assuming the 
latest hardware in the user-end means narrowing the market 
down. It is self-evident that publishing it on a thin wire 
network, like the Internet, requires a lot more consideration. 
This is one reason why researchers put effort in trying to find 
out functional methods to reduce the amount of data to be 
transmitted. 
In this paper, we will look closely into one approach in making 
the transfer of visual data more feasible for a thin wire network. 
This approach is called "foveation"; it has been successfully 
utilized as a compression method for 2D images meantime it 
has a claim that it is a more “natural” way for humans to sec 
things. Similar approaches, exploring or including the same 
idea, have been called “perceptually-driven” or “gaze- 
contingent” by different researchers. 
2. HUMAN VISUAL SYSTEM AND COMPUTER 
GENERATED VISUALIZATION 
“In order to modulate the LOD of an object based upon its 
perceptual content, one must first possess an appreciation for 
the way in which the human visual system is designed and how 
it is believed to function.” (Reddy 1997) 
The term “foveation” comes from the word “fovea”, which is 
the name for a region inside the human eye. The description of 
the phenomenon was given in many sources, in two recent 
notes, it is as follows: 
“The human visual system implements an elegant comprise 
between the competing goals of maximizing field of view, 
maximizing spatial resolution, and minimizing neural resources: 
It encodes a large field of view using a retina with variable 
spatial resolution, and then when necessary, uses high-speed 
eye movements to direct the highest-resolution region of the 
retina (the fovea) at specific points in the visual scene.” (Perry 
& Geisler, 2002) 
“Resolution is much greater at the fovea in the center of vision 
than in the periphery. The foveal area subtends about two 
degrees of visual angle, i.e. a disc about two cm in diameter 
held at arms length. If we move ten degrees off to the side, the 
amount of fine detail that can be perceived declines by a factor 
of ten. It is also the case that color vision mostly takes place in 
the fovea. However, the periphery of vision is very important to 
our sense of orientation and to our sense of self motion." (Ware, 
2003) 
As can be seen in the Figure 1, in the next column the visual 
acuity distribution is symmetrical towards the temple and 
towards the nose. 
Figure 2 illustrates the foveation over an example image. 
Figure 2: Exaggerated visual illustration of foveal vision. Here 
is a 6-fold (6 levels of detail) foveation from the point of gaze, 
or in other words, 6 levels of details. Keep in mind that this is a 
very rough representation, compared to the 35-fold that the 
human eye can achieve. 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV. Part BS. Istanbul 2004 
  
  
Blindspot 
b 
Relative Visual Acuity 
e 
Cn 
  
  
Degrees from Fovea 
Figure 1: The human visual acuity distribution. 
Normally it is stated that human eyes have roughly about a 35- 
fold difference (35 levels of detail) between the fovea and the 
periphery (Nakayama, 1990). The following figure shows an 
incremental use of blur function to reconstruct the scene 
employing foveation (Figure 2).