1
General / Re: "can't compress image" texturing error with EXR
« on: May 01, 2022, 12:38:35 PM »
Forgot to say, was building a 16K texture file.
Just tried with a 4K, and it worked.
Just tried with a 4K, and it worked.
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.
Pages: [1] 2
2
General / "can't compress image" texturing error with EXR
« on: May 01, 2022, 12:36:25 PM »
Hi,
I was attempting to build a texture using 32bit OpenEXR.
All looked well until the end when it gave me the message "can't compress image".
And no texture file was produced
I've not managed to find this on google or the forums.
Any idea what the problem is.
Metashape pro 1.7.5 Build 13229
mac os
iMac i9, 40GB RAM
I was attempting to build a texture using 32bit OpenEXR.
All looked well until the end when it gave me the message "can't compress image".
And no texture file was produced
I've not managed to find this on google or the forums.
Any idea what the problem is.
Metashape pro 1.7.5 Build 13229
mac os
iMac i9, 40GB RAM
3
Bug Reports / Re: Mask defocus area works then un-works
« on: December 30, 2021, 08:13:38 PM »
Hi Alexey,
Yes, I had just been doing one or two.
Just tried it on all cameras and that seemed to work. So yes.
Thanks.
Yes, I had just been doing one or two.
Just tried it on all cameras and that seemed to work. So yes.
Thanks.
4
Bug Reports / Mask defocus area works then un-works
« on: December 30, 2021, 02:00:20 PM »
Hi.
After not having. much luck with the generate masks command, I've tried again and watching more closely.
Except.
When I run the command, I will see a mask appear on the image, that looks right. Before it then disappears and is replaced by a mask that follows only the outside bounds of the mesh. Not the focus area that was initially selected.
Changing the defocus area vlaue, will change the shape of the initial mask that appears briefly, but not the version that it leaves me at the end.
I've tried values from 0.1 to 4 with the same result.
Has anyone else encountered this, have I missed a setting or is it strange behaviour?
After not having. much luck with the generate masks command, I've tried again and watching more closely.
Except.
When I run the command, I will see a mask appear on the image, that looks right. Before it then disappears and is replaced by a mask that follows only the outside bounds of the mesh. Not the focus area that was initially selected.
Changing the defocus area vlaue, will change the shape of the initial mask that appears briefly, but not the version that it leaves me at the end.
I've tried values from 0.1 to 4 with the same result.
Has anyone else encountered this, have I missed a setting or is it strange behaviour?
5
General / Re: "nan" markers and 0 paramaters with image variant calibraiton
« on: December 30, 2021, 01:11:54 AM »
Hi Alexey,
No. This is all with projects that have already aligned successfully (with the image-varient options selected during alignment).
But on reopening the project (s, i'm finding it in any project i've tried the image variant option) the F has 0 when I come back to it.
When I turn off the image-variant option following your instructions, the F and other columns disappear - as expected.
When I turn them back on, either there is no F column, only cy, cx, k etc.. or no extra columns at all.
but the project will work still work (non 'nan' values).
Other times if I turn image-variant options back-on, or on for the first time for a calibration group, it will have all the columns and all the values will be the same as each other before i optomize.
It feels quite confusing. Hope you can understand.
No. This is all with projects that have already aligned successfully (with the image-varient options selected during alignment).
But on reopening the project (s, i'm finding it in any project i've tried the image variant option) the F has 0 when I come back to it.
When I turn off the image-variant option following your instructions, the F and other columns disappear - as expected.
When I turn them back on, either there is no F column, only cy, cx, k etc.. or no extra columns at all.
but the project will work still work (non 'nan' values).
Other times if I turn image-variant options back-on, or on for the first time for a calibration group, it will have all the columns and all the values will be the same as each other before i optomize.
It feels quite confusing. Hope you can understand.
6
General / Re: "nan" markers and 0 paramaters with image variant calibraiton
« on: December 29, 2021, 09:51:20 PM »
In the parent project I first noticed this I'm having similar issues, but instead of the columns starting at cx, I don't have any columns after date time.
Hope this helps point out what it/ or I is getting wrong.
Thanks again.
Hope this helps point out what it/ or I is getting wrong.
Thanks again.
7
General / Re: "nan" markers and 0 paramaters with image variant calibraiton
« on: December 29, 2021, 08:57:15 PM »
Thanks
I did
1. open camera calibraiton (all values are 0)
2. turn of image variant paramters
3. closed camera calibration
4. open camera calibration - it had numbers but after date time it had Cx, Cy. and all have the same numbers
5. close camera calibraiton with ok
6. set optomize cameras and set into the queue....
then metashape crashed on a different project. - before optomize was run (i think)
so I opened the file again (unopotomized i think), and do 1. - now after date and time it does has F. and all the numbers are different (not 0)
then I repeat 2 - 4. Same Cx, Cy straight after date time.
then i closed the calibration and opened again, this time I have the same, but if I do it with the other camera group I get the F column too.
Thanks
I did
1. open camera calibraiton (all values are 0)
2. turn of image variant paramters
3. closed camera calibration
4. open camera calibration - it had numbers but after date time it had Cx, Cy. and all have the same numbers
5. close camera calibraiton with ok
6. set optomize cameras and set into the queue....
then metashape crashed on a different project. - before optomize was run (i think)
so I opened the file again (unopotomized i think), and do 1. - now after date and time it does has F. and all the numbers are different (not 0)
then I repeat 2 - 4. Same Cx, Cy straight after date time.
then i closed the calibration and opened again, this time I have the same, but if I do it with the other camera group I get the F column too.
Thanks
8
General / Re: "nan" markers and 0 paramaters with image variant calibraiton
« on: December 29, 2021, 11:25:17 AM »
Thanks, will do.
Just found the problem reappear too. Don't think I've done anything other than build mesh and texture, close the file and open (run on batch) after correcting the problem, so very odd that it's come back.
Just found the problem reappear too. Don't think I've done anything other than build mesh and texture, close the file and open (run on batch) after correcting the problem, so very odd that it's come back.
9
General / Re: "nan" markers and 0 paramaters with image variant calibraiton
« on: December 29, 2021, 01:44:35 AM »
Hi,
I'm afraid i've fixed the issue by resetting alignment then re-running workflow-align photos (without reset alignment).
I can still send the project if you think it might have something diagnositic in it?
Thanks
I'm afraid i've fixed the issue by resetting alignment then re-running workflow-align photos (without reset alignment).
I can still send the project if you think it might have something diagnositic in it?
Thanks
10
General / "nan" markers and 0 paramaters with image variant calibraiton
« on: December 28, 2021, 11:57:44 AM »
Hi,
I've not used the image variant paramters much before and seem to have something odd happen.
I have a series of chunks through time with markers for position between them.
2 camera groups for each chunk both starting with a precalibration values from an earlier good model with the same camera (but different working distance), one with a fixed focus, the other variable focus (not enough room to stick to the focus)
when i first ran everything it seemed ok, but at some point i found a lot of "nan" in my markers. I eventually tracked this down to my variable calibration group parameters all estimated to 0.
I'm not sure when this happened as it all looked like it worked first time.
Is there something that's likely to lead to this situation (having adaptive camera fitting on, or off during alignment for example)
Thanks
I've not used the image variant paramters much before and seem to have something odd happen.
I have a series of chunks through time with markers for position between them.
2 camera groups for each chunk both starting with a precalibration values from an earlier good model with the same camera (but different working distance), one with a fixed focus, the other variable focus (not enough room to stick to the focus)
when i first ran everything it seemed ok, but at some point i found a lot of "nan" in my markers. I eventually tracked this down to my variable calibration group parameters all estimated to 0.
I'm not sure when this happened as it all looked like it worked first time.
Is there something that's likely to lead to this situation (having adaptive camera fitting on, or off during alignment for example)
Thanks
11
General / Re: Defocus masks
« on: July 29, 2021, 07:00:47 PM »
Nope, I take it back, there is something strange.
Every so often, and more often than I'd like the mask is the exact opposite of what it should be.
I think my camera positions are good? The geometry is nice and clean and all other things (placing markers etc) appear to work as expected. But not the generate masks.
Is this just a mistake in my camera alignment?
Every so often, and more often than I'd like the mask is the exact opposite of what it should be.
I think my camera positions are good? The geometry is nice and clean and all other things (placing markers etc) appear to work as expected. But not the generate masks.
Is this just a mistake in my camera alignment?
12
General / Re: Defocus masks
« on: July 13, 2021, 09:56:35 AM »
Actually, ignore this, I was misunderstanding the result. I think it works as expected when one knows what to expect! Good tool, thanks!
13
General / Re: Defocus masks
« on: July 10, 2021, 01:32:01 PM »
Hi, yes I've had something similar.
I've found objects that don't have very distinctive surface details get masked out completely while some of the supporting turntable furniture (scales etc) will get selected even though they are way out of focus.
If I'm using a fixed focal distance, is there a way I could manually tell metashape the focused area in focus in one image it can then apply to all the other images?
Would starting with a model-based mask help with this perhaps?
I've found objects that don't have very distinctive surface details get masked out completely while some of the supporting turntable furniture (scales etc) will get selected even though they are way out of focus.
If I'm using a fixed focal distance, is there a way I could manually tell metashape the focused area in focus in one image it can then apply to all the other images?
Would starting with a model-based mask help with this perhaps?
14
Bug Reports / Re: crash with GPU 1.5.5 imac catalina
« on: January 14, 2020, 02:46:37 PM »
Thanks,
sorry for delay, I've updated to 1.6 and so far all is well.
sorry for delay, I've updated to 1.6 and so far all is well.
15
Bug Reports / crash with GPU 1.5.5 imac catalina
« on: December 28, 2019, 12:52:26 PM »
Updated OS to catalina and metashape to 1.5.5
crashes when building dense cloud/mesh from depths maps but only on larger files -> i've got it to work on a ~40 photo file, but 300, and 1400 photo files it kills the imac completely - i.e. next time i see it's at the log in screen so I don't have and agisoft crash report, only apple.
it will plod through merrily enough if i turn the GPU off though.
27" 2019 iMac with 8GB radeon vega 48 and 40GB ram
crashes when building dense cloud/mesh from depths maps but only on larger files -> i've got it to work on a ~40 photo file, but 300, and 1400 photo files it kills the imac completely - i.e. next time i see it's at the log in screen so I don't have and agisoft crash report, only apple.
it will plod through merrily enough if i turn the GPU off though.
27" 2019 iMac with 8GB radeon vega 48 and 40GB ram
Pages: [1] 2
