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 - holgar

Pages: [1] 2
General / Error: Index Overflow
« on: February 26, 2020, 10:14:35 AM »
Hello, I am trying to build a Mesh from a 300 Million Points Denseclould. After some hours I get a popup saying: Index overflow. The Console also says: ERROR: Index Overflow.
Any Idea what that is and how to solve it? I have 128GB Memory and thought I can handle this amount. Is it a memory problem?

General / Workflow Suggestion
« on: February 24, 2020, 10:45:03 AM »
my current Workflow is:
1. Load Images
2. Align
3. Gradual Selection and Optimize Cameras
4. Denscloud
5. Mesh

This workflow allways require me to interact manually to make the gradual Selection and Optimize Camera. I would like to have this step in the Batchprocessing. So I could start a Job and the next day have a finished Mesh. This would save a lot of time as right now it takes me normally two days to get a finished mesh as in the morning I have to manually step in for the gradual Selection.

I Capture Reality for example you just set a reprojection Error threshold before you start. Than you just press run and the next day you have a finished Mesh.

General / Turntable and Alignment Failling too often
« on: February 21, 2020, 01:45:06 PM »
I am using Agisoft Metashape/Photoscan since many years. But I have allways had troubles to align my Images I make on my turntable. I take a lot of care to produce good images:
- Checking for diffuse Light
- Checking for plain background
- Checking for good sharpness and resolution
- Checking good overlap (100-200 Images per Object)
- I even use Crosspolarization techniques to prevent reflections.
- On small objects I sometimes use focus stacking (works sometimes. Sometimes not. Capture Reality works allways)

No matter what I do the Alignment fails very often on many images. Many times it is half the Cameras of one side (when the object rotated 180 degree.).
This is my main Problem with Metashape. I have tried Capture reality a few times with my failing Objects. And Capture reallity allways does a perfect Job aligning.
Also Capture Reality produces way more Tie Points than Metashape ( I set Tiepointlimit in Metashape allways to Zero).

So I wonder if there is something that can be improved in Metashape to make aligment more stable? I dont want to use masks (which helps sometimes). This adds an extra step that I want to avoid. 
I would really like to keep using Metashape as I like it a lot. Also the pricing of Capture Reality prevents me from using it for my private Projects. Also I like some features and the Interface of Metashape more.
Please let us know if there is anything that can/will be improved in this regard.

Thanks a lot for the great work and the frequent updates. Greetings, Holger

General / Reason for limiting Tiepoints?
« on: September 28, 2018, 04:51:08 PM »
Hello, There is one thing I wonder. Why Limiting Tiepoints? I set my Tiepoint limit to "0" to have no Limit in all my Projects.
Is there any issues with that? What would be a case where I want to set a limit? Why is it set not to "0" by default?
greetings, Holger

General / AMD Cards support on Mac
« on: March 08, 2017, 12:46:27 PM »
is there a plan to fix the issue with AMD Cards? I have my MacPro here which was a heavy investment and cant use Photoscan 1.3. Thats a serious issue. I would like to know if there is plans to fix the issue to make further purchase decisions.

greetings, Holger

General / RMS Reprojection Error vs. Max reprojection Error
« on: June 20, 2016, 11:00:03 AM »
Hello, I made scan with 577 Cameras. I did as always some filtering and applied a optimizes Cameras.
Now I have a much less RMS reprojection Error. But the Max reprojection Error got worse.
I wonder what is the reason here. Should I prefer the optimized version with less RMS reprojection or should I keep the unoptimized version with less Max Reprojection Error?

General / Bigger Reprojection Error after Optimization
« on: May 16, 2016, 11:39:13 AM »
I have another Issue. After Doing the Reprojection Error Filtering (gradual selection) I run the Optimize Cameras. The Problem is that the Reprojection Error is much higher after that. It goes from 0.2 to 1.9!!
What can cause that? Should I better not do the optimizes step than?

greetings, Holger

General / Gradual Selection: Image count selects almost all points
« on: May 16, 2016, 11:20:09 AM »
I have a questions about the Image Count filter in gradual selection. I make photos of objects at the size of a bottle. I use many images....100-200 21 MP images with a fullframe camera.
I get fairly high number of matching points (around 300.000). Anytime I try to use the Image Count filter of 2 almost all Points get selected. Like 250.000 of 300.000. Is that normal? I guess its not good to filter that many points for optimization.
What am I doing wrong? Do I need even more overlap? I do Turntable Fotos which I mask. and I take about 36 Images per round.

Any Help would be very welcome.
greetings, Holger

General / Which Optimize Settings (without GCP)
« on: November 20, 2015, 11:22:33 PM »
Hi, I am generating 3D models without any GPC (I have Standardversion). So my Workflow is that I filter the unacurate points from the SparseCloud with gradual selection and than I do Optimize Camera.
Now I am not sure which settings I need there. I have read that k4 should be used for wide lenses. I use 50-70mm so I leave that option off. Also I am shooting at a Studio-Lighning indoor and I have very sharp Images to  begin with.

But what about the other Parameter? In which cases do I need which parameters. Is it possible to explain this without technical language but with practical Scenarios?

General / Dealing with DOF
« on: February 13, 2015, 02:27:27 PM »
Hello, me again with another Question. I hope I can some day give you some valluable Infos on my experiences back ;-).

I would like to scan small Objects. I am considering buying a Canon EF 50mm/ 2,5/ Makro.
Now the Problem with Marco is certainly the shallow DOF. So I wonder how PS is dealing with blurred parts of the images.

I have made a test with proccessing Focus-stacks with a Software called Helicon Focus. The results where very good in PS. But the Process of taking Focusstack Photos is very long. I would prefer to instead take more Photos from different angles and distances.

If I take enough pictures from different angles I assume I will still get good results even with only small parts in Focus.

Another though I had was to make masks based on Sharpness of the images (Photoshop has such a fucntion). Would that help in PS?

My question is how Ps is dealing with blurred parts? Will big unsharp parts of the image reduce quality? Or are thoose part ignored in Recontruction?

General / Reuse Depth Maps
« on: February 13, 2015, 01:04:05 AM »
there is this option in the preferences to keep depth maps. Also there is the Option in the Denscloud generation Dialog "reuse depth maps".

I wonder how I should use these Options and in which cases I can benefit from these options? I have not found any helpfull info in the documentation.

greetings, Holger

General / Sparse Cloud: very few Points
« on: February 12, 2015, 02:36:18 PM »
Hello Agisoft Community,
I have found a pretty cool Object which I want to Scan. It is a burned out Carwreck.
So I waited for the best weather conditions with a cloudy sky and took the photos.
I made 700 Images with 21 Mpixel with my Canon 5DII.

The Images are very sharp and have good quality. Now I run the Alignment and all Cameras aligned fine.
I used the default settings for Keypointlimit of 40k. This morning I checked the result and I am suprprised I got only 12K points in my sparse clould. After doing some filtering I ended up with 1700 Points.

I would have expected much more points with 700 Cameras. Each camera should find 40K Points.

Before I start the Densecloud calculation I would like to ask you if I should try another Alginment with a higher Keypointlimit.  I fear my sparsecloud is way to small.

General / Halfres-Image and Ultra vs. Fullres-Image and High setting
« on: February 11, 2015, 03:17:49 PM »
Hello, I am wondering if anybody has compared Processing time and results with these two Scenarios:

Full-Res image but with High-setting (densecloud)


Halfres-Image (size reduction before importing in PS) and instead Ultra-High setting in Denscloud.

I wonder if performancewise and Qualitywise it makes any Difference. As far as I understood the High-setting downsamples the original image by half. So from my understanding I could also reduce the Resolution before I load into PS. But does that have any advantages? I am wondering if anybody has tested that already?

greetings, Holger

General / taking Photos of a Building: Right Shooting Order
« on: February 10, 2015, 03:24:48 PM »
I am planning to do scanning of big buildings and Big Areas like Archological Sites.
Now I am wondering how I should plan to take the pictures as it is not as simple as the Examples in the Documentation.
I am not doing UAV. I shoot everything from tripod.
If I have a Complex Building. Is there a prefered order in which I should take the photos?

If I have to do a Complex Wall which is lets say L-Shaped. Would you circle the Wall completly Row by Row? Or better split the Wall into pieces and make single Segments of the Wall one by one?

As an simple example...a big wall. Should I take the Photos in lines from left to right?
Which of the following methods would you do:

General / Optimizing Cameraalignment for better Results
« on: April 11, 2014, 10:19:14 AM »
I am using the Standardversion of PS to make scans of small objects. I want to get as much detail as possible. But using UltraHigh setting produces much noise. I take my Photos in a Studio with a lot of Light. I make 18 Megapixel Photos with a very good lens. So the Image quality I can not improve.
I have read that in the Pro  Version of PhotoScan there is a Option for Alignmentoptimization. Would that help me to get rid of the noisy surfaces. Or are there other Ways of optimizing in the Standard Version?

I usually get reproduction Error of 0.2-0.3
Effective Overlap is 2.68.
After analyzing the Imagequality I get Results ranging from 0.7-0.9
are that good Vallues?  should I take more Images?

Thank you for any Help. PS is such a great software and I am happy I found it.
Greetings, Holger

Pages: [1] 2