In this article, we will explore in detail how to best take full advantage of the wide array of functionalities offered by our repurposing tools.

Reddit Spoofer is compatible with jpeg, jpg, gif, mp4 and png. Simply drag or select the files you would like to process and you will see a screen like this.

Let's break down what each of these options mean:

Label is the text that will be appended to the beginning of the name of the output file.

So if your original file is named "IMG7072" and you type the label "Reddit", the name of the outputs will be "Reddit_(round number)_IMG7072". Very useful for keeping files organized.

Effect Intensity refers to the amount of visual effect that will be applied to each file (brightness, contrast, etc). If you want the best possible results then pick either Medium or High.

Available only when spoofing GIFs, you are able to toggle between Default (which keeps the quality of the original GIF as high as possible) vs FFmpeg (which is slightly faster, but at a cost in GIF quality).

Rounds refer to how many times you want the original file to be spoofed.

Here you can change the location the files will be saved. By default, it will save the spoofed files to the same directory as the original file.

As it is the case with all of our tools, you can always apply these settings to all selected files by clicking on this icon:

Here is the Reddit Spoofer in action:
​

Currently, the UI for the TikTok Spoofer is identical to the Reddit Spoofer, so refer to the text above for instructions on how to use it.

Obs: TikTok Spoofer is currently in beta, so make sure to use the Detector after spoofing (aim for as close to 20% as possible). We are working on an improved TikTok Spoofer that will be released soon.

DA Spoofer is compatible with images only. The images must have a face present in them, or it will not work (for images without a face present in them, use the Reddit Spoofer at max settings).

Simply drag or select the files you would like to process and you will see a screen like this.

Let's break all of the functionalities down.

Works exactly the same as described in the Reddit Spoofer section.

This section refers to the intensity of the Adversarial Attack that will be conducted on the target image.

So, what is an Adversarial Attack?

An adversarial attack is like tricking a smart computer by giving it slightly changed data it wasn't expecting. Imagine you show a clear picture of a cat to a computer, and it correctly says, "That's a cat." But if you make tiny, almost invisible changes to the picture, the computer might now say, "That's a dog," even though it still looks like a cat to human eyes. These tiny changes are the "adversarial attack," confusing the computer into making a mistake.

How is this relevant to DAs?

Our DA Spoofer basically tricks their system into thinking your face (or your client's face) is someone else's, effectively spoofing pictures.

What mode should i use?

"Low" with EXIF, Resizing, Pixel Attack and Brightness is good enough for DAs. Anything beyond that is not generally necessary.

!!Warning!!: DA Spoofer is heavily reliant on your CPU, and processing times will depend on how good your CPU is. From our testing with a modern MacBook (Apple silicon chip), you can spoof one file on "Low" in less than 1 minute. Medium will increase the processing time significantly. High will increase it even further. There is nothing that can be done from our end until hardware starts catching up with the latest AI systems, as you are effectively running a neural engine locally.

Again, Medium and High are for advanced users and not necessary in 99% of cases.

Applies a random resize effect to each spoofed image.

Applies random brightness to each spoofed image.

Applies random hue effect to each spoofed image.

Applies a Pixel Attack to each spoofed image.

Adds random geolocation EXIF metadata to each spoofed image.

Here is the DA Spoofer in action: