Training files missing

[ADD-eNavarro]

Hi! I'm trying your great project, but when building the project there's a bunch of lines like this one:
"1>C:\WINDOWS\Microsoft.NET\Framework\v4.0.30319\Microsoft.WinFx.targets(712,5): error BG1002: File 'Depth_ML\data\nyu2_train\bedroom_0047_out\95.png' cannot be found."
And yes, in the training folder "bedroom_0047_out" a lot of images are missing (53 pairs of 95, precisely).
Since probably the training would be better with all the images, could you please provide the missing ones?
Merci!

Also, if I remove the missing pictures from the CSV and the project, I get an error:

=============== Running AutoML experiment ===============
#########################################################
Running AutoML regression experiment...
Press any key to stop the experiment run...
The thread 0x2228 has exited with code 0 (0x0).
The thread 0x38e4 has exited with code 0 (0x0).
The thread 0x2c10 has exited with code 0 (0x0).
The thread 0x33b0 has exited with code 0 (0x0).
The thread 0x35f0 has exited with code 0 (0x0).
The thread 0x182c has exited with code 0 (0x0).
The thread 0x20ec has exited with code 0 (0x0).
The thread 0x389c has exited with code 0 (0x0).
The thread 0x3854 has exited with code 0 (0x0).
Exception thrown: 'System.ArgumentException' in Microsoft.ML.AutoML.dll
Exception thrown: 'System.Reflection.TargetInvocationException' in mscorlib.dll
Exception thrown: 'System.Reflection.TargetInvocationException' in mscorlib.dll
Exception thrown: 'System.Xaml.XamlObjectWriterException' in System.Xaml.dll
Exception thrown: 'System.Windows.Markup.XamlParseException' in PresentationFramework.dll
'The invocation of the constructor on type 'Depth_Estimation_ML.MainWindow' that matches the specified binding constraints threw an exception.' Line number '6' and line position '9'.

[Titibo26]

Hi, the problem is probably a missing file or wrong path to the file. Try replacing all image path and names with existing ones.
For the exception, i don't think it is related to missing files, it's a bad usage from my side of autoML.

this project was inconclusive so I abandoned it and i can't help you on it, sorry.

If you are interested in depth prediction with monocular cam, you can check this project : https://github.com/intel-isl/MiDaS
You can run ONNX model in .Net. Results were not perfect for me, but at least you can work with it.