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.


Topics - Paulo

Pages: [1] 2
1
General / X Y Z definition in Slave offset for camera rig
« on: June 24, 2019, 07:13:51 AM »
Hello all,

I am processing a multi band camera rig (Altum) data set with Metashape. In camera calibration pane, I have blue band camera as Master and want to adjust location w.r.t. to this reference. I allow adjust location and get offset of -0.15 m in Z for camera 6 LWIR w.r.t. to Master Blue camera 1....see attachment....

If I understand X,Y,Z definition for these offsets to be Y in direction of flight, X to right of flight direction and Z up, then this would mean that in general camera 6 would be 15 cm lower than camera 1 .... however when I print the adjusted  camera coordinates, I get camera 6 to be around 15 cm higher than camera 1....

Can you explain? ???

2
General / Multi band imagery processing
« on: June 19, 2019, 06:50:15 PM »
I am processing a 6 band imagery data set from Altum....

As one of the outputs, I create a NDVI index map using raster calculator with output band 4 as (B5 - B3) / (B5 + B3).

So index values vary from 0 to 1...

But when I move the cursor over output NDVI map, sometimes I get values in 29 000 to 31 000 range displayed in bottom right of screen which correspond to output band 5 LWIR...

Why is this so?

3
General / Plan Mission
« on: June 15, 2019, 10:02:56 PM »
I have a suggestion.... many times we may have a project done with a fixed wing drone and would like to  plan a mission in order to build a 3d model of a structure part of the project.

In camera model parameter of Plan Mission, it would be useful if we could select or define a camera different from one used previously.... i.e. one that would be used by the quad copter to capture the  structure of interest...

In example, original project was captured with an eBee fixed wing equipped with a Canon IXUS 125HS  camera.

4
Python Scripting / Number of Cameras in version 1.5.2 vs 1.4.5 API
« on: April 04, 2019, 09:54:58 AM »
Hello,

I am using a script that loops over all cameras in a chunk to extract information... This script works well in 1.4.5 and 1.5.2... However I am noticing sometimes that if I open a PhotoScan project in 1.5.2 then the API detects more cameras than actually present in project....

For example I have a project with 118 cameras and if in 1.4.5 I do following command len(chunk.cameras), I get correct result, 118....
However, opening same project in 1.5.2, the same command returns 168 instead of 118... It adds 50 empty cameras to chunk....

Is this a bug^?

Please see attached logs for 1.4.5 and 1.5.2 versions...

5
General / Processing times version 1.5.2 vs 1.4
« on: March 24, 2019, 01:22:05 PM »
Hello all,

i am comparing processing a full body model using 120 photos in versions 1.5.2 and previous 1.4.5..... on same computer

2019-03-24 02:56:47 Platform: Windows
2019-03-24 02:56:47 CPU: Intel(R) Core(TM) i7-3630QM CPU @ 2.40GHz (laptop)
2019-03-24 02:56:47 CPU family: 6 model: 58 signature: 306A9h
2019-03-24 02:56:47 RAM: 16.0 GB
2019-03-24 02:56:49 OpenGL Vendor: NVIDIA Corporation
2019-03-24 02:56:49 OpenGL Renderer: GeForce GTX 670MX/PCIe/SSE2
2019-03-24 02:56:49 OpenGL Version: 4.6.0 NVIDIA 417.01

I am generating a mesh from depth maps in high quality with strict volumetric masks on….

For version 1.4.5 the time to generate  mesh from depth maps is 49 min.

For version 1.5.2 the time to generate  mesh from depth maps  is  1 h 42 min.

I realize the mesh from depth maps algorithms have changed but the considerable increase in processing time is concerning….

Hoping to get your thoughts on this issue….

Attached are processing logs from both versions and screen shot of both models side by side….

Seems model 1.52 is a little better but substancial increase in processing time does not seem to justify such marginal quality improvements….

Note that in both cases following tweak was used in preferences/advanced :  main/dense_cloud_max_neighbors  60

6
Python Scripting / Script using strict volumetric masking
« on: March 23, 2019, 11:57:37 PM »
Hello,

in 1.4 version, I was using a script to generate model from depth maps with high quality and strict volumetric masking on…

The code line was:

chunk.buildModel(surface=PhotoScan.Arbitrary, interpolation=PhotoScan.EnabledInterpolation,  source=PhotoScan.DepthMapsData, quality=PhotoScan.HighQuality, volumetric_masks=True, reuse_depth = True, keep_depth = True, vertex_colors = False)

Now in 1.5.2, I get invalid keywords for quality and reuse_depth parameters….

Why are these parameters no longer available? These keywords were avalable in 1.5.0 API reference...

7
General / Planning Motion tool
« on: February 04, 2019, 04:22:34 PM »
Hello Agisoft support,

I am curious to know what exactly does the new tool Planning motion (replacing Optimize coverage) do in MS 1.51....


Im trying it on a 118 UAV 16 Mpix image set at 4.8 cm GSD with double grid coverage and it is taking an enormous time ( 17 hours and counting...)


Attached Screen captures....


More insight on this tool and its various input parameters (which I left at default value) will be appreciated!



8
Python Scripting / requests Module fails to import in Metashape 1.5
« on: February 02, 2019, 01:38:16 AM »
Hi all,

I am trying to run a python script that ran correctly on PS1.4...

the script contains following code:

import requests

and get following error in Metashape

2019-02-01 07:21:23   File "C:/Users/paul.pelletier/Documents/Pi3d/AutoProcess_v1/AutoProcess_v15_adapFit_Newmesh.py", line 14, in <module>
2019-02-01 07:21:23     import requests
2019-02-01 07:21:23 ImportError: No module named 'requests'
2019-02-01 07:21:23 Error: No module named 'requests'

How can I get request module to work in Metashape?

I found solution to this problem consulting the Agisoft solutions page https://agisoft.freshdesk.com/support/solutions

and using following system command:

"%programfiles%\Agisoft\Metashape Pro\python\python.exe" -m pip install requests

9
General / Importing dense cloud
« on: July 03, 2018, 12:36:03 AM »
Hello,

I am importing a LAZ dense cloud into PS 1.4.2 and the process imports the 360 Million points in about 25 min but then  it switches to Calculating normals ….  and this part has been running for an hour without much progress….

Why is it so time consuming?

I am running on a i7-3630QM CPU with 16 GB of Ram


Any help is appreciated...


Is is now running for more than 8 hours and Calculating normals… is stuck at 99%

Will it ever conclude?

10
Python Scripting / Reset Mask in Python API vers. 1.4
« on: November 28, 2017, 09:01:15 PM »
Hello ,

here is my problem... in version 1.4, there is a new mesh generation using strict volumetric masking.

I want to use this in my current workflow using script python.

In my current workflow, I mask all my 100 images from background images so as to expedite the alignment phase.

Now I want to use experimental mesh creation method with strict volumetric masking using

chunk.buildModel(surface=PhotoScan.Arbitrary, interpolation=PhotoScan.EnabledInterpolation,  source=PhotoScan.DepthMapsData, quality=PhotoScan.HighQuality, volumetric_masks=True, vertex_colors = False)

However from blog i see that for this I do not need all images masked, so I need to reset masks for most images and just keep maybe 3 images masked  before doing the new mesh method....

How can I do this with Python API?

11
Python Scripting / LoadReference error
« on: March 17, 2017, 11:59:19 AM »
Hello,

I am writing a script version 1.2.6 that would load reference from a text file (tab delimited) with nxyzabc information using following code

chunk.loadReference(file, 'csv' , 'nxyzabc',  '\t', group_delimiters=False, skip_rows=1)

I get following encoding error in console:

2017-03-17 02:52:52   File "C:/Users/paul.pelletier/Documents/Photoscan/Importeoypr_files_chunks.py", line 29
2017-03-17 02:52:52 SyntaxError: Non-UTF-8 code starting with '\xa0' in file C:/Users/paul.pelletier/Documents/Photoscan/Importeoypr_files_chunks.py on line 29, but no encoding declared; see http://python.org/dev/peps/pep-0263/ for details

Any help?

12
Bug Reports / GPU processing failed on Spherical image project
« on: February 15, 2017, 05:38:35 AM »
Hello,

I am processing a project consisting of 8 spherical images in PS 1.3.

In Image alignment using 0 key point limit and 0 tp limit, I get an error due to GPU.

So I deselect GPU in Preferences and redo Image Alignment with no problem.

Before doing Dense cloud, I reactivate GPU and get errors that is unable to use GPU processing and switching to CPU and final error at end.

Attached is log file ...

Hope you can help here,


13
General / Importing Precise camera intrinsics and extrinsics
« on: January 25, 2017, 07:13:08 AM »
Hello all,

I have a xml file containing calculated Exterior orientation (Extrinsic) and interior orientation (intrinsic) for a set of photos.

If I import this camera file to the chunk containing said photos, my idea was that we could then proceed to Dense cloud Generation as camera positions and camera parameters are known from imported xml.

Then, why is the program asking us to go over the sparse cloud generation (key point detection ,matching, etc...).?

Any insight will be appreciated!

14
General / Versionn 1.3 Added image based shape drawing support
« on: December 16, 2016, 05:39:04 AM »
Dear support,

The image based shape drawing support is great in version 1.3.

But I have a problem relating to productivity relating to capturing long line strings in an image.

For example I am capturing a wall. I place first 2 vertices in a photo but when I need to move in image to continue capturing following vertices of feature by pressing V or Space key, the shape is placed with just 2 first vertices. Thus I would need to place another line string to continue representing this wall. Would it be possible to switch between navigation mode and line placement mode without having line being terminated at each switch.


Thanks for your help,

15
General / No overlapping images error
« on: July 19, 2016, 10:23:27 PM »
Hello all,

I am trying to correct an Ortho I built from a DEM which inturn was generated from a Dense Cloud of high quality.

I draw a polygon around an area to correct and when I right click and select Assign Images, I get

No overlappings images found errror message.

See attached Screen Copy...

Any ideas why is this occuring?

Latest version of PS 1.2.5 2735 used....

Pages: [1] 2