process Point Of Interest (POI) images with Photoscan

[mfran2002]

Hello everybody,
I've to elaborate my images taken during the POI procedure with Photoscan.

I've never done it before...

During the fly I have taken the images and then I have processed them in the same usual way:

1. align
2. optimization
3. build dense cloud
4. mesh
5, texture

but I get the result in the attached image

I don't think it's good: where is my error? may someone help me?

My only idea is the "surface" value in the mesh elaboration: I put "HeightField" but maybe I should put "Arbitrary"...could it be?

THANKS!

[Alexey Pasumansky]

Hello mfran2002,

Yes, you need to use Arbitrary surface type, since Height Field option creates 2.5D surface. But please not that mesh generation in Arbitrary more is considerably more resource demanding operation compared to Height Field method.

[mfran2002]

great Alexey!
thank you again!

and what about buildUV mapping parameter?
have I to use AdaptiveOrthophotoMapping value?


doc.chunk.buildUV(mapping = PhotoScan.AdaptiveOrthophotoMapping)

[Alexey Pasumansky]

Hello mfran2002,

You can use either Adaptive Orthophoto or Generic, because common Orthophoto mapping will not produce texture for vertical surfaces.

[mfran2002]

fantastic Alexey!

thank you very much!!!

see picture...