Exception followed by queue full about frigate

[blakeblackshear]

The mask works by looking at the bottom center of the bounding box for the detected person in the image. If that pixel in the mask is a black pixel, it ignores it as a false positive. In my mask, the grass and driveway visible from my backdoor camera are white. The garage doors, sky, and trees (anywhere it would be impossible for a person to stand) are black.

@jon102034050 If you just omit the mask: property in your yaml config, it should generate an all white one for you on the fly.

@drcrimzon The only reason that line could fail is if OpenCV was unable to read your mask file. Looking at the docs, this is what it says:

If the image cannot be read (because of missing file, improper permissions, unsupported or invalid format), the function returns an empty matrix

I have been saving my file out of Photoshop as a BMP with these settings:

[blakeblackshear]

Thanks. I will fix that in the next release.

[JonGilmore]

I had this same stack trace. Does your mask file exist for all of your cameras?

see this issue: #29

[drcrimzon]

My masks are there, is there a specific bitmap format they are supposed to be in? I just created a WxH sized image in gimp and exported as bmp. I verified permissions, and triple checked config, it looks correct.

[drcrimzon]

Well, I recreated the masks with a gray scale bitmap, which dropped the size from 7MB to 2MB, this solved the crash it appears.

[drcrimzon]

Well, it crashed again with the 2MB bitmaps, I created 500k ones using imagemagick:
convert -size 2560x1440 -colors 254 xc:white mask.bmp, which looks like it's working so far.

Not sure if it's still an issue, If I remove the mask from the config, it looks like the code is supposed to create one on the fly, but that generates an error currently.

I'll let this run over night and see if it is really fixed, and then take a look at the auto create tomorrow.

[JonGilmore]

fwiw, my mask is literally just a white bmp, and its around 4.5MB and I've had no issues. I'll be perfectly honest though and say that I don't know much about how the masks work, and it doesn't seem to be really well documented (yet).

[drcrimzon]

Those are the settings, I was using. The container ran overnight with the new bmp i generated with convert.

I also tried omitting mask, but that resulted in an error trying to execute np.zeros in the generation of the mask. I'll try it again and post the exact error.

[drcrimzon]

@blakeblackshear, I went back and removed the mask from my config. The error was np not defined. Checked and video.py is missing numpy.
Added
import numpy as np and rebuilt, and it resolved the auto mask creation.