Messages

16

Feature Requests / Clarify/complete reference preselection

« on: July 24, 2023, 11:46:08 AM »

Hi,

Alignments are sometimes giving pretty hard times, especially when a short deadline is given for the project (almost every time...). It would be nice to add "advanced settings" to the images preselection, like:
- Generic: groups size
- Reference / Estimated: Neighbors distance, number and/or max h offset (would be pretty useful for interiors with multiple floors)
- Groups, links (like a folder by room and folders with images linking different spaces)

In fact, those processes can't always be automatic and usually the user knows more informations than the software. Hence, it's absolutely necessary to give users the access to certain parameters they know exists. Grouping them in "advanced settings" can warn the user to know exactly what he expects, or leave default settings. If some of those functions already exists as tweaks, then I ask you one more time to give a list of the available tweaks! Agisoft is supposed to be a professionnal software, so please stop considering that users will break the software every time you give them some room on parameters. Those advanced settings can be part of the professionnal version only for example.

Thanks

17

Feature Requests / Upgrade on Tile Map Service export for non-WGS84 projects

« on: June 30, 2023, 01:29:30 PM »

Hi,

It would be awesome to generate an index file within the Tile Map Service orthomosaic exports which can handle data with CRS other than WGS84. Leaflet or other open source xyz APIs could easily handle that.

Thanks!

18

General / Rolling shutter compensation

« on: June 29, 2023, 04:20:36 PM »

Hi,

Using a precalibrated system, I realised that the rolling shutter has a huge impact on final camera positions. Knowing accurate speed, heading, orientation and position for each frame and the sensor readout time, is there a way to help the software to get more accurate? Position already helps a lot, but sometimes alignment still goes wrong.

Thanks!

19

General / Re: Stability of depth maps calculation

« on: December 20, 2022, 02:29:38 PM »

Hello,

After more tests, I confirm that the filtering time is equal or even lower (up to 20%) on CPU (1.8.4). Knowing that it is the operation that causes GPU crashes, is it possible to have a Python script to better adjust the use of the hardware?
Which means, a command to start the creation of depth maps, then one to deactivate the GPUs and a last one to filter the depth maps? I think the software's functions should be subdivided at this level but the functions are not public because they won't make anything alone.

This would be extremely useful, because the first step takes 5 to 10 times more time on CPU! I really don't want to decompile the software to modify it by myself...

Thanks in advance !

20

Bug Reports / Re: Panoramas are cropped around 45 degrees down

« on: December 16, 2022, 05:50:51 PM »

In case it helps anyone, the problem appeared on version 1.8.3. On version 1.8.4, the problem is solved.

21

Bug Reports / Re: Panoramas are cropped around 45 degrees down

« on: December 16, 2022, 05:49:14 PM »

Hello,

I tried many things this afternoon, then I realized that the machine used still had version 1.8.3. So I updated to 1.8.4 and the issue seems to be fixed.

In any case, it would be nice to detail a little more the bug fixes applied between versions in the changelog, even if this case was not one of them. In this case I had looked if anything had been changed between 1.8.4 and 2.0.0, but if it had been marked somewhere lower I would have seen it.

Thank you anyway for your feedback. I did not find how to delete the duplicate post in "general", but I will specify in response to the latter that on version 1.8.4 the problem was not noticed.

22

Bug Reports / Lot of mismatches with 14bit marker n° 63

« on: December 16, 2022, 05:14:37 PM »

I'm using 14 bit markers and there is a lot of mismatches with some markers under number 100. By chance I'm not using them but I have to delete all those mismatches everytime using markers.

The most problematic marker is the number 63, but there are problems with markers 16, 19 and others too for example.

A workaround to fix this issue would be appreciated. Thanks!

23

Bug Reports / Panoramas are cropped around 45 degrees down

« on: December 16, 2022, 12:41:08 PM »

When I generate panoramic images, although the preview shows me that my images cover the ground up to plus or minus 10 degrees above nadir, the generated image is cut off at around 45 degrees above nadir (it's indeed filled up with nothing, export is in format 2:1).

This is very annoying because the images are correctly aligned and this area also contains correct tie points. The tripod and other technical elements are visible and not part of the scene, but it does not matter because for analysis purposes I need to see this area.

Is there a solution to handle this problem? I specify that I cannot use the "setup boundaries" function because in any case it will not solve my problem (and it outputs the same cropped image). I have to generate a lot of panoramas and this option is not available in the batch process tool. I take this opportunity to also say that the layout, name and availability of settings between batch and standard use should be harmonized! Another example is that it is sometimes possible to export data in precise CRS via the batch but not via the standard functions... It is always divination...

Anyway I have to export images in 2:1 format, so impossible to output cropped images, I need the full area.
Thanks in advance !

24

Bug Reports / Panoramas are cropped around 45 degrees down

« on: December 14, 2022, 07:06:02 PM »

When I generate panoramic images, although the preview shows me that my images cover the ground up to plus or minus 10 degrees above nadir, the generated image is cut off at around 45 degrees above nadir (it's indeed filled up with nothing, export is in format 2:1).

This is very annoying because the images are correctly aligned and this area also contains correct tie points. The tripod and other technical elements are visible and not part of the scene, but it does not matter because for analysis purposes I need to see this area.

Is there a solution to handle this problem? I specify that I cannot use the "setup boundaries" function because in any case it will not solve my problem (and it outputs the same cropped image). I have to generate a lot of panoramas and this option is not available in the batch process tool. I take this opportunity to also say that the layout, name and availability of settings between batch and standard use should be harmonized! Another example is that it is sometimes possible to export data in precise CRS via the batch but not via the standard functions... It is always divination...

Anyway I have to export images in 2:1 format, so impossible to output cropped images, I need the full area.
Thanks in advance !

25

General / Re: Very poor and bad documentation

« on: December 07, 2022, 11:21:46 AM »

Hi Arie,

I mean, on big and challenging projects where you have to find your way to properly assemble you data, the documentation can't help.

Of course, as you said, there is a lot of informations on the helpdesk, social media groups and forum. All those informations should be packed in the documentation. I'm sure that after years working with Metashape you can use it like a charm, but I've never been so lost starting with a software. The worst thing is that many things have changed over time, they are not named the same and it makes the helpdesk, social media groups and forum incomplete or wrong.

The worst case is when you try to use the Python API, you'll find most of the solutions on the forum but once again, some function's name have changed. It wouldn't be very hard in Agisoft's side to concatenate all solutions given as examples in the python API doc, so that we won't have to search in the forum for ages. Most modern software have an online documentation structure which allow you to search efficiently trough the documentation and other discussion platforms at once. It's not the best example, but QGIS has a good documentation: https://docs.qgis.org/3.16/en/docs/index.html
You can choose the version, make efficient search and you know what? It's open source, so they could just have told users to look at the user manual, helpdesk, social media groups and forum to solve their problem but they made a clean documentation instead!

26

General / Re: Stability of depth maps calculation

« on: December 07, 2022, 11:02:37 AM »

Hi,

For the log files there could be a little issue where you activate the log, then Metashape crashes and the log is no more activated (need a clean close).

It's probably a matter of IO limit, but I can't do better overall, I'm working with a RAID 0 array of two M.2 SSD (around 6Go/s). It's then possible that the CPU has a faster access to data than the GPUs.

For the repeated crashes, there is always hardware issues, but as I run huge processes without any problem on other programs without any problem, it's just a problem of error management in your side. As an example, when you're writing a code to control something wireless, you'll get a lot of hardware problems (mainly loss of connection). But you have to deal with it and build something stable, which handles errors without systematically crashing and without rejecting the fault on the components..

27

General / Re: Can’t set mesh CRS in empty project

« on: December 05, 2022, 04:00:51 PM »

So it seems that the CRS can't be properly defined in an empty chunk...

Thanks for the script, I'll keep it, but for now it's way easier to import any e57 cloud and then import the mesh in the right CRS.

It's a bit weird because once you have imported a dense cloud, you can then import a mesh and choose the CRS.

28

General / Re: Stability of depth maps calculation

« on: December 05, 2022, 03:48:45 PM »

Hi,

I'll try to export all this data next time I get this problem. By the way, it would be nice to just set a directory to save logs, rather than manually set a file before processing. Usually I save the log after 2 or 3 crashes but it's a lot of debug data lost.

I was very surprised to see that the CPU performed as good as a dual GPU on this step! It could be because I was processing in lowest quality but it's still good to know as this step crashed only on GPUs. We always compare softwares to their performances, but stability is way more important. Nevertheless, I'm a Python developer too and I know how hard it is to work with GPUs...

29

General / Re: List of tweaks

« on: December 05, 2022, 03:38:08 PM »

Hi,

I'm talking about general tweaks which could optimize processes in certain cases, like dense_cloud_max_neighbors for example. You can signically reduce process time knowing you have far less than 100 neighbors per images (especially when shooting interiors, it's really uncommon to have more than 10 neighbors).

I think it's better to keep this tweak system, so that users won't break the software. Users who know what they do would be able to find a tweak corresponding to their needs without crawling trough the whole forum. And as far as I know, the name of some tweaks have changed, so old posts often don't work.

A simple list, with all "public" tweaks and a brief description would do the job and it would be very appreciated!

30

General / Re: Wide angle lens for full frame Nikon mirrorless camera

« on: December 05, 2022, 03:18:01 PM »

Not sure if works well with a GoPro, it depends of the kind of project you're working on. Generally, higher resolution sensors mean more matches and the overall result is better (so it's better to have 100 50Mpx images than 400 12.5 Mpx images for the same area and resolution). I made some tests using the import video features, but even with a good camera you'll never get better images than shooting RAW. Distorsion on wide angle lenses is really not a problem. You even get better results with a low end wide angle than using a pricey telephoto lens in photogrammetry. The only application where you could use a focal higher than 100mm is for panoramic views.