NudeNet - automatic censoring AI software

[Miss_Leah.]

When I first ran the detector, it did download the model file and put it in

Code: Select all

.NudeNet/detector_v2_default_checkpoint.onnx

in my home directory.

Does that file exist? Should be about 140MB.

Doesn't seem so, no, in fact I don't seem to have a NudeNet folder in my home directory at all, or maybe I'm looking in the wrong place?

[kerkersklave]

Which operating system are you using?

[Miss_Leah.]

Which operating system are you using?

MacOS

[kerkersklave]

Hm, I believe it should be /Users/<yourusername>/.NudeNet

Try opening a console and do

Code: Select all

cd ~/.NudeNet

The . is important and the directory might be hidden in a file browser.

[Miss_Leah.]

Hm, I believe it should be /Users/<yourusername>/.NudeNet

Try opening a console and do

Code: Select all

cd ~/.NudeNet

The . is important and the directory might be hidden in a file browser.

You're quite correct, the file Is there, so that shouldn't be the issue

Code: Select all

classes					detector_v2_default_checkpoint.onnx
classifier_model.onnx

[Miss_Leah.]

I downloaded a specific version of python for this, 3.7.7, could that have an impact? It was the one listed in the guide

[kerkersklave]

Can your run

Code: Select all

md5 -r detector_v2_default_checkpoint.onnx

and paste me the result?

Just to check, that you really have the same file and it didn't get corrupted during download or something.
The stack trace clearly says, that this is commig from instatiating the model from that file. So something has to go wrong there...

[Miss_Leah.]

Can your run

Code: Select all

md5 -r detector_v2_default_checkpoint.onnx

and paste me the result?

Just to check, that you really have the same file and it didn't get corrupted during download or something.
The stack trace clearly says, that this is commig from instatiating the model from that file. So something has to go wrong there...

Code: Select all

d41d8cd98f00b204e9800998ecf8427e detector_v2_default_checkpoint.onnx

[kerkersklave]

The file seems to be different from mine. Could you rename it and run the example again? Should be downloaded again.

The python version should be fine and it is not a python issue, the error comes from somewhere in native code, if this does not help, I have to look deeper into what it means.

[Miss_Leah.]

The file seems to be different from mine. Could you rename it and run the example again? Should be downloaded again.

The python version should be fine and it is not a python issue, the error comes from somewhere in native code, if this does not help, I have to look deeper into what it means.

Woop! It worked! Thank you so much!! One final question now, and I won't bother you any more, how do you choose categories for censoring?

[kerkersklave]

Always glad to help fellow kinksters to create new content, looking forward to see what you're making.

You can pass a list of labels to censor() like:

Code: Select all

detector.censor(
    './test.jpeg',
    out_path='./out.jpeg',
    parts_to_blur=['EXPOSED_BREAST_F', 'EXPOSED_GENITALIA_F']
)

[Miss_Leah.]

Always glad to help fellow kinksters to create new content, looking forward to see what you're making.

You can pass a list of labels to censor() like:

Code: Select all

detector.censor(
    './test.jpeg',
    out_path='./out.jpeg',
    parts_to_blur=['EXPOSED_BREAST_F', 'EXPOSED_GENITALIA_F']
)

My hero! Thank you so much

[willow11]

Well you threw me into a Rabbit hole with this.

I was a total beginner with python this morning.

• 1. i wanted to go through a folder and censor all the images in it and put it out to another folder. Managed to do that

• 2. Add an option to replace the black censor box with a gaussian blur. Done

• 3. Add an option to replace the black censor box with a pixilation box. Done

Done with this for now. Now should i share this somewhere?

[forbiddendesire]

Well you threw me into a Rabbit hole with this.

I was a total beginner with python this morning.

• 1. i wanted to go through a folder and censor all the images in it and put it out to another folder. Managed to do that

• 2. Add an option to replace the black censor box with a gaussian blur. Done

• 3. Add an option to replace the black censor box with a pixilation box. Done

Done with this for now. Now should i share this somewhere?

That's impressive, I had a cursory glance at the github and docs for this and it seemed quite a bit difficult to do, if you're a beginner with python that's quite amazing. I'm going to try downloading it and see if I can figure it out myself

[willow11]

I just realized this seemed like i wanted to show off. It was more like: I got obsessed with this and spent hours on google.

But i know my way around R so that helped. Now I need some material to censor any suggestions?

Btw. you can specify the parts that gets censored. Ofc i wanted all the cocks and faces to be visible and only censor the girl parts like breasts and genitalia. I did it like this

ptb=['EXPOSED_ANUS','EXPOSED_BUTTOCKS','EXPOSED_BREAST_F','EXPOSED_GENITALIA_F','COVERED_BREAST_F','EXPOSED_BELLY']

detector.censor(
*image path*,
out_path=*image output path*,
visualize=False,
parts_to_blur=ptb)


these are the options from the readme

**Default Detector classes:**
|class name | Description |
|--------|:-------------------------------------:
|EXPOSED_ANUS | Exposed Anus; Any gender |
|EXPOSED_ARMPITS | Exposed Armpits; Any gender |
|COVERED_BELLY | Provocative, but covered Belly; Any gender |
|EXPOSED_BELLY | Exposed Belly; Any gender |
|COVERED_BUTTOCKS | Provocative, but covered Buttocks; Any gender |
|EXPOSED_BUTTOCKS | Exposed Buttocks; Any gender |
|FACE_F | Female Face|
|FACE_M | Male Face|
|COVERED_FEET |Covered Feet; Any gender |
|EXPOSED_FEET | Exposed Feet; Any gender|
|COVERED_BREAST_F | Provocative, but covered Breast; Female |
|EXPOSED_BREAST_F | Exposed Breast; Female |
|COVERED_GENITALIA_F |Provocative, but covered Genitalia; Female|
|EXPOSED_GENITALIA_F |Exposed Genitalia; Female |
|EXPOSED_BREAST_M |Exposed Breast; Male |
|EXPOSED_GENITALIA_M |Exposed Genitalia; Male |


If you got any questions just ask. Cheers