Hello people,
I am developing a application similar to google earth.Currently to keep
it
simple I am only using satellite images and no terrain data.Satellite
images
are stored in a quadtree structure on the disk.
So my questions are
1) How to determine the level of detail of satellite images?.For sure we
need to use the camera altitude above MSL as the input so that when we are
close to earth surface images are detailed and when we are high enough
then images are less detailed...
Second thing which i feel should be taken as input is the camera pitch
so that even if we are close to earth surface and the pitch is such that
camera's view vector is at a grazing angle with the earth surface then
level of detail of image should be less(coarser).
So is there any paper which describes this or some article which
explains this?For time being I am using a crude function and am looking
for
a better one.
2) Can anyone give me inputs for the pre-ca****ng/pre-loading of data for a
google earth kind of application or for a flight sim kind of app where the
direction a user may move is random.What I think is that depending
on the movement vector a new camera position should be extrapolated along
with camera's attibutes and this new camera position should be
used to pre-load the data...But if someone is having better thoughts or
some
paper/article describing it, then it would be a great to read it.
Thanks in advance.....
Regards
Sandeep.
