Messages

1

Bug Reports / Re: Tips for crash at end of align photos?

« on: January 20, 2015, 04:05:14 PM »

Sorry to waste your time Alexey 

I thought I did all the updates for my Win 7 Ultimate, but now today the SP1 appeared...

Everything works perfectly now!

Thank you very much!

2

Bug Reports / Re: Tips for crash at end of align photos?

« on: January 20, 2015, 01:47:27 PM »

Hello Alexey,

Yes, both computers have the latest windows updates and nvidia drivers.

Thanks!

3

Bug Reports / Re: Tips for crash at end of align photos?

« on: January 20, 2015, 01:00:04 PM »

...but my other pc has exactly the same CPU.

Well I hope you can fix it with a different setting. I would like to use the network processing.

Thanks again!

4

Bug Reports / Re: Tips for crash at end of align photos?

« on: January 20, 2015, 01:15:09 AM »

Hi Alexey,

Indeed I tried to remove the entire folder after uninstall, but it didn't help.

My CPU model is i7-4820K 3.7 GHz.

Thank you!

5

Bug Reports / Tips for crash at end of align photos?

« on: January 19, 2015, 11:10:43 PM »

Hi!

Photoscan Pro 1.1.0.2004 keeps crashing on 1 of my computers in the last step of Align Photos, when estimating camera positions (see attached Windows crash report).

The strange thing is that I have two other computers on which Photoscan runs fine. One of them is almost an exact copy of the first, but was built later.

So I updated every driver I could think of, including the latest OpenGL. I also had Python 2.7 installed for some other project, and uninstalled that. I uninstalled Photoscan and deleted the directory. But nothing seems to help.

I'm sure it's a small software thing, but I can't figure it out.

Does anyone have some more tips on things I can try?

Thanks!
Fernando

6

Python and Java API / Re: How to set region volume to 2 camera locations?

« on: July 10, 2014, 06:57:01 PM »

Thank you so much Alexey! I'll give this a try.

7

Python and Java API / Re: How to set region volume to 2 camera locations?

« on: July 09, 2014, 05:05:05 PM »

Hello Alexey,

No, that is not necessary.

Thanks

8

Python and Java API / Re: How to set region volume to 2 camera locations?

« on: July 09, 2014, 02:19:17 PM »

Hi, maybe someone could help me? Or is this a really dumb question? Sorry, I'm a total Python noob 

9

Python and Java API / How to set region volume to 2 camera locations?

« on: July 07, 2014, 01:39:21 AM »

Hi, I was wondering if I could manually set the region bounding box to 2 points in space: 2 camera locations.

So for example these are the camera locations:

cam1(0,0,0)
cam2(100,100,100)

Together they form the maximum / most outer points of the region volume.

How can I set the region to that volume?

Thanks!

10

Python and Java API / Re: Is there a Python equivalent of Tools > Import Masks?

« on: June 23, 2014, 08:15:52 PM »

Great, thanks!

11

Python and Java API / Is there a Python equivalent of Tools > Import Masks?

« on: June 22, 2014, 04:48:32 PM »

Hi, I have images with an alpha channel, and I was wondering if there is a way to create masks from those alpha channels with Python?

Thanks!

12

Python and Java API / Re: How to use color tuple in createDifferenceMask?

« on: June 22, 2014, 03:11:40 PM »

Thank you Alexey. I still don't know what's wrong with my own script, but I have a working version now!

13

Python and Java API / Re: How to use color tuple in createDifferenceMask?

« on: June 22, 2014, 02:39:39 PM »

Hi, I'm using version 1.0.3.

14

Python and Java API / How to use color tuple in createDifferenceMask?

« on: June 22, 2014, 02:11:33 PM »

Hi, I've tried this:

photo_mask = utils.createDifferenceMask(original_photo, (200, 200, 200), tolerance=50, fit_colors=True)

but Python returns an error:

RuntimeError: invalid color format

What am I doing wrong?

Thanks!

15

General / Re: Much better results with 0.9.1 version than with 1.0 ...?

« on: January 25, 2014, 01:18:25 PM »

Hi Eduardo,

That is a fair request, see the attached comparison.

I've set the newest version from this week to the maximum values (except I kept dense cloud to 'high').

Even with more points matched in the 'align photos' phase, the results are worse.

If I hadn't used 'extrapolated' there would be a gap on the side of the nose.

I would expect to get the same results or better.

@Agisoft: have your base algorithms changed that much?

Thanks,
Fernando