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 - paulm

Pages: [1]
1
General / Re: Thermal Orthomosaic with FLIR DUO PRO R 336
« on: May 31, 2018, 04:59:06 PM »
Have you looked at Chapter 3 of the manual (http://www.agisoft.com/pdf/photoscan-pro_1_4_en.pdf)? That should provide at least some general information on how to approach your task...

2
General / Re: CPU multi-core questions
« on: May 31, 2018, 03:22:11 PM »
Hi Alexey,

I just sent the dataset as a .psz file. This contains the results of doing Workflow -> Align Photos from the GUI. The "Estimating camera locations" step used only 1 CPU core in this case.

Paul

3
General / Re: CPU multi-core questions
« on: May 31, 2018, 11:49:06 AM »
So I did some further testing and it seems I DO get multi-threaded computations when processing the same dataset on my laptop (4 cores with HT + Quadro M1000M), while on a more powerful compute node (16 cores + 2x Tesla K40m) I once again see the single-core computations. I use a Python script to make sure I'm doing the exact same processing on these two nodes and the single-core part is due to chunk.alignCameras().

The matchPhotos() step is roughly 2.5 times faster on the compute node, but after that performance switches to single-core.

4
General / Re: CPU multi-core questions
« on: May 30, 2018, 05:30:31 PM »
The step is "Estimating camera locations", sorry for misremebering the name. The scene has 199 photos of 4000x3000 pixels, also containing 25 markers, no masks. Settings: accuracy = high, generic preselection enabled, reference preselection enabled, 40k key point limit, 4k tie point limit, adaptive camera model fitting.

The highest I'm seeing is around 108% CPU out of a possible 1600% (i.e. 16 cores). The node has 32 GB RAM and also 2 Tesla K40m's.

5
General / Re: Viewing a 3d Model in VR
« on: May 30, 2018, 05:15:03 PM »
Quote
The model is 1.5gb.

How large is it in number of polygons?

In VR, especially with say a Vive headset, you need really beefy graphics performance as the headsets run at 90-120 frames per second and less will cause stutter in the VR experience. Displaying in stereo (left/right eye) potentially halves the framerate, but modern VR viewers can use tricks to make the performance hit much smaller than a factor of 2.

In practice, the model might need to be simplified quite a bit to be smoothly viewable in VR.

Quote
I need to know the best format to export out. And which software to view the model in.

These two are related, as the viewer you're going to use will dictate what kind of format to export in (or need to convert to). Good general 3D model formats are Autodesk FBX (binary, lots of features), Collada (XML-based), Wavefront OBJ (verbose as it is text-based) or PLY (binary, but does not handle textured materials).

Sketchfab.com has a VR-viewer, but it requires you to upload the model to their site, which might not be what you want. You can use Unity3D, but that requires you to do some scene building (and even some programming). The Vive comes with an application called Workshop that allows to add and view 3D models, but it's a bit convoluted (https://developer.valvesoftware.com/wiki/SteamVR/Environments/Photogrammetry)

 I actually don't know any good standalone VR viewers for 3D models.

6
General / CPU multi-core questions
« on: May 30, 2018, 04:33:00 PM »
Hi,

I'm trying to debug some performance issues between different nodes. I'm seeing the camera alignment phase of the Align Photos step only using 1 CPU core (i.e. no mult-threaded and/or GPU computations). It this expected behaviour? I can't seem to find any documentation on this topic. This is with 1.4.2 Pro under Linux.

Secondly, are there any performance limitations in demo mode? Due to having only a single professional license I am running one test in activated mode on one node and another in demo mode on a different node.

Regards,
Paul

7
Python and Java API / Re: Writing log to file
« on: May 30, 2018, 12:51:08 PM »
I would be interested in this as well. Setting log_enable to True and setting a file name in log_path doesn't seem to be enough. And is there an API call to add log messages to the application log from a user script? E.g. app.log("....")

8
General / License no longer usable?
« on: May 11, 2018, 12:35:04 PM »
Hi,

It has been a while since I last used our copy of PhotoScan Pro. On the host where it is installed I now get the error "Wrong host for license (-4)". When I try to reactivate using the license code I have I get the error "Activation key is already in use". I've tried looking on other hosts where we have run PhotoScan in the past, but I don't see any activated copies there.  How do I get out of this situation? Is there any way to tell from the license file or license code on which host it was last active (so I can deactivate it there)?

Regards,
Paul

9
Bug Reports / TIF file doesn't open correctly
« on: January 11, 2017, 12:01:38 PM »
I converted a movie to individual .tif frames, using ffmpeg (ffmpeg -i <movie> -vf fps=1 frames%04d.tif). The resulting tif files are not correctly read by PhotoScan Pro, even though they open fine in GIMP and tools like tiffinfo. Once loaded in Photoscan the images show a corrupt image, in grey scale. On the console warnings of the form "PackBitsDecode: Not enough data for scanline 193." are displayed.

I can provide an showing the problem.

Pages: [1]