Processing images failed

I was trying to process a set of 84 images using a fresh install of Web ODM v1.9.11 build 45, on a new Dell Computer. The computer has 32gb of ram and an intel I7 processor.

I have tried 3 times to process these images with no luck. I have also tried another set of 76 images and could not process them either.

The images were shot at 250AGL, with 80x80 overlap using a Mavic Pro with 12mp camera. The link to the images is: Dropbox - _0279B - Simplify your life

The Log from the Task folder in Web ODM is in the dropbox folder as well since it to too large to copy and paste below.

Thanks in advance for any help.

2 Likes

Where in the process does it fail?

What settings have you tried?

2 Likes

I am using default settings. Here is part of the log that shows the errors -

===== Dumping Info for Geeks (developers need this to fix bugs) =====
Child returned 3221225477
Traceback (most recent call last):
File “C:\WebODM\resources\app\apps\ODM\stages\odm_app.py”, line 94, in execute
self.first_stage.run()
File “C:\WebODM\resources\app\apps\ODM\opendm\types.py”, line 346, in run
self.next_stage.run(outputs)
File “C:\WebODM\resources\app\apps\ODM\opendm\types.py”, line 346, in run
self.next_stage.run(outputs)
File “C:\WebODM\resources\app\apps\ODM\opendm\types.py”, line 346, in run
self.next_stage.run(outputs)
[Previous line repeated 3 more times]
File “C:\WebODM\resources\app\apps\ODM\opendm\types.py”, line 327, in run
self.process(self.args, outputs)
File “C:\WebODM\resources\app\apps\ODM\stages\odm_meshing.py”, line 24, in process
mesh.screened_poisson_reconstruction(tree.filtered_point_cloud,
File “C:\WebODM\resources\app\apps\ODM\opendm\mesh.py”, line 207, in screened_poisson_reconstruction
system.run(’"{reconstructmesh}" -i “{infile}” ’
File “C:\WebODM\resources\app\apps\ODM\opendm\system.py”, line 106, in run
raise SubprocessException(“Child returned {}”.format(retcode), retcode)
opendm.system.SubprocessException: Child returned 3221225477

I just successfully ran the same images on my old laptop running WebODM V 1.9.2, Build 28. Took a little over 35 minutes on a Dell laptop with 16 gig of ram, and I7 processor.

Still can’t figure out why Build 45 won’t process these images.

1 Like

My laptop has the Intel iris plus graphics

1 Like

Welcome! Sorry you’re having this trouble!

I think we hit on a bit of a regression when we enabled CUDA-accelerated processing in more phases of our processing pipeline.

I’ve created an issue on our GitHub to track this and other similar errors that have been cropping up this past week after Build 44 and Build 45.

1 Like

Thank you. What was the latest stable version that worked well? Can you send me a link to that version so I can process images while the issue on build 45 is resolved.

1 Like

We don’t have a policy of keeping superseded releases, and I don’t have any archived at the moment.

In light of the trouble some folks have been having, we might revise that.

Could you do me a favor and verify that you can process the Brighton Beach dataset with defaults?

1 Like

If you want to use an earlier version of the processing engine (ODM) you can as a workaround download one of the releases from: Releases · OpenDroneMap/ODM · GitHub

And install ODM in C:\WebODM\app\resources\apps\ODM (replacing the existing directory).

That will keep the latest WebODM UI, but use an earlier processing engine version.

2 Likes

Processed perfectly fine locally with my NVIDIA GPU.

Created on: 11/30/2021, 5:11:56 PM
Processing Node: node-odm-1 (manual)
Options: auto-boundary: true, crop: 0, debug: true, dem-gapfill-steps: 4, dem-resolution: 1, dsm: true, dtm: true, feature-type: orb, matcher-neighbors: 16, mesh-size: 300000, min-num-features: 15000, orthophoto-resolution: 1, pc-classify: true, pc-filter: 1, pc-geometric: true, pc-quality: high, pc-sample: 0.01, resize-to: -1, use-3dmesh: true, verbose: true
Average GSD: 2.42 cm
Area: 40,286.77 m²
Reconstructed Points: 10,334,819

jvhooser-9623-report.pdf (3.0 MB)

I’m going to pull that out to go back to my Intel HD Graphics 530 iGPU from my i7-6700k and I’ll try to re-run the dataset again.

1 Like

Failed to process with my NVIDIA GPU uninstalled and running from my Intel HD Graphics 530.

Created on: 11/30/2021, 6:16:17 PM
Processing Node: node-odm-1 (manual)
Options: auto-boundary: true, crop: 0, debug: true, dem-gapfill-steps: 4, dem-resolution: 1, dsm: true, dtm: true, feature-type: orb, matcher-neighbors: 16, mesh-size: 300000, min-num-features: 15000, orthophoto-resolution: 1, pc-classify: true, pc-filter: 1, pc-geometric: true, pc-quality: high, pc-sample: 0.01, resize-to: -1, use-3dmesh: true, verbose: true

image

3 Likes

Perfect test case.

3 Likes

I tried the Brighton Beach set and processing failed. Here is a link to the log

3 Likes

Welp :frowning_face: I really want to fix this bug because it’s obviously affecting many users, but none of the machines I can test with seem to exhibit this bug, which really seems hardware dependent (or perhaps less likely due to software environment differences). So I’m still unable to reproduce this.

If somebody has this problem and is OK lending me their machine remotely so that I may install some development tools and run OpenMVS in debug mode to track this issue, please let me know. :pray:

2 Likes

Doesn’t Saijin_Naib work for you? :smiley:

4 Likes

Yeah :stuck_out_tongue:

Maybe reproducing on my potato doesn’t count :rofl:

2 Likes

Somehow I missed the part on this thread where he was able to trigger the bug :grinning_face_with_smiling_eyes:

3 Likes

Lol. That explains it.

2 Likes

My post when Piero was reading:
homer-simpson-homer
This is the real explanation…

4 Likes

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.