Bart Rammeloo Posted August 17, 2003 Share Posted August 17, 2003 Perhaps a strange question, but I have the impression that VW interpolates in a rather linear way between 3D loci. Is this the case, or is it reality a cubic interpolation type? The reason I ask this, is because, when I use a higher polycount, this doesn't result in curving slopes between 3D-loci, while I would expect to have at least some kind of more visible smoothing effect between sampling points? Can anyone clarify? Thanks in advance. Kind regards, BaRa. [ 08-17-2003, 06:54 PM: Message edited by: BaRa ] Quote Link to comment
Petri Sakkinen Posted August 18, 2003 Share Posted August 18, 2003 As most DTMs, that of VW works out contours from the TIN-model (triangulated irregulard node) which 'simply' (the algorithms I have seen are anything but simple) connects the dots (3D loci or 3D poly vertices). What happens between points, is not known and linear interpolation is just as valid as any other. For the human brain (at least that of an experienced landscape analyst), it is possible to 'see' (=guess) the 'trends' of landforms, based on what is 'above' and 'below' any contour, but I'm not sure if I would like to have a stupid computer making these intuitive guesses - after all, the computer has not made any geomorphological field trips, done orienteering or tried to find wild mushrooms with a topographic map. So, to make the point: any other interpolation would try to 'add' something to the data. A good surveyor takes his/her spot levels in an intelligent manner and that certainly helps. Terrain modeling is not an exact science in any case. And now for something completely different: in addition to TIN models, there are also DTM engines based on the raster concept. With a raster (grid) it is indeed easy to traverse the data set along grid lines / raster cells and do 'trend' type calculations, more complex imterpolations / extrapolations and even 'typology' analyses. This, however, requires huge data sets, compared with the TIN approach. Whoa - I try to sound like an expert! That I am not. Definitely not. Quote Link to comment
Bart Rammeloo Posted August 18, 2003 Author Share Posted August 18, 2003 Thanks Petri, you kinda confirmed what I suspected. Pitty though that there's no option to use other interpolation types. I know this would have nothing to do with the real site situation, but it would make nicer 3D-models Cheers, BaRa Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.