Forum

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - cstal

Pages: [1]
1
Python and Java API / Re: issue with Photoscan.Model
« on: November 25, 2013, 08:47:51 PM »
Dear Mr. Pasumansky,
Thank you very much for the suggestion. It seems to work now!

2
Python and Java API / Re: issue with Photoscan.Model
« on: October 04, 2013, 09:07:08 AM »
Hello,
I would like to give a short update: all data in the project has been transformed from the Belgium coordinate system to UTM31N coordinates, using WGS84 (EPSG::32631). When I run the script again, the loaded models (also converted to UTM) are still not correctly positioned. This procedure did not solve the problem. Do you have a solution yet?
Thanks again!

3
Python and Java API / Re: issue with Photoscan.Model
« on: September 25, 2013, 09:05:11 AM »
Dear mister Pasumansky,
The project contains a large number of airborne images with known orientation- and rotation parameters. I'm using the Belgian national coordinate system for this (EPSG:31370), which is explicitly defined in the Ground Control settings.
The idea is to import a third party model, texture it and export it again:

Code: [Select]
model = PhotoScan.Model()
model.load("D:/modelPath_IN/modelName.obj", "obj")
doc.activeChunk.model = model

doc.activeChunk.buildTexture(mapping="adaptive orthophoto", blending="average", size=1000, count=1)
model = doc.activeChunk.model
model.save("D:/modelPath_OUT/modelName.wrl", format="vrml", texture_format="jpg", export_texture=True)

It seems that the model is loaded, but the export returns a black model, transformed in a very strange coordinate system. I double checked the coordinates of the obj-file to be imported, and they are correct. Moreover, when I perform this procedure manually via Tools -> Import -> Import Mesh and after selecting my coordinate system, it works fine.

Re-assigning the (same) coordinate system using Python by
Code: [Select]
cs = PhotoScan.CoordinateSystem()
cs.init("EPSG:31370")
doc.activeChunk.projection = cs
puts the project in a weird viewpoint and still does not load the models correctly.

I hope you have a solution.

Thanks again for your assistance

4
Python and Java API / Re: issue with Photoscan.Model
« on: September 24, 2013, 06:15:38 PM »
Dear,
The automated model loading still positions my models in a very weird coordinate system. Is there already a way to properly import models in the same coordinate system as the active chunk?
I'm using the Agisoft Pro 1.0.0.
Many thanks in advance!

Pages: [1]