[judge2020]

Well, according to this site:

> Perhaps there is a reason that they don't want really technical people looking at PhotoDNA. Microsoft says that the "PhotoDNA hash is not reversible". That's not true. PhotoDNA hashes can be projected into a 26x26 grayscale image that is only a little blurry. 26x26 is larger than most desktop icons; it's enough detail to recognize people and objects. Reversing a PhotoDNA hash is no more complicated than solving a 26x26 Sudoku puzzle; a task well-suited for computers.

> I have a whitepaper about PhotoDNA that I have privately circulated to NCMEC, ICMEC (NCMEC's international counterpart), a few ICACs, a few tech vendors, and Microsoft. The few who provided feedback were very concerned about PhotoDNA's limitations that the paper calls out. I have not made my whitepaper public because it describes how to reverse the algorithm (including pseudocode). If someone were to release code that reverses NCMEC hashes into pictures, then everyone in possession of NCMEC's PhotoDNA hashes would be in possession of child pornography.

https://www.hackerfactor.com/blog/index.php?/archives/929-On...

[JimDabell]

> 26x26 is larger than most desktop icons; it's enough detail to recognize people and objects.

This doesn’t seem to be true. These are 26×26 icons, for instance:

https://cdn.dribbble.com/users/975594/screenshots/6193663/ic...

[TonyTrapp]

Depends on what you define as "Desktop Icons". Many icons in Windows are 16x16. The icons that literally sit on the Windows desktop are 32x32 or bigger, though.

[nomel]

Here's a greyscale 26x26 of Obama:

torso: https://imgur.com/a/Oz0pnyF

closeup of face: https://imgur.com/a/Hr1OS2P

I guess it has to be decided if wide access to blurry images like this is better or worse than finding full resolution images. This all assumes there's not some obfuscation layer, I suppose (is that possible?).

[judge2020]

We might also see DL image upscaling used. Obtain similar-looking pictures (at 128x128 or even smaller), downscale to 26x26, train DL to do 128x128 or larger upscaling using the original picture as training data. Assuming most content in the actual DB is similar, you can probably obtain plausible results. A similar technique is used by Peacemaker Filmworks to upscale 4k to 8k: https://youtu.be/umyglbDr4IE?t=257

[nomel]

That will be an incredibly questionable dataset used to train it. At that point, there would probably be no need to upscale these images. Just downscale then use this questionable DL to re-upscale any image.