They go over the effects of compression - which they say only degrades the protection - but at the same time also degrades the identification accuracy of the AI model.
So if your crappy 6 megapixel camera cannot take a clear shot of the cloaked pixels - or effectively applying a blur filter - would also affect the AI detection.
They are lucky in this case because compression does defeat the cloak to some degree— However, compression at those levels also defeats the ability of Facebook to accurate identify you.
Very interesting! Great timing for myself, as I just pulled the Chez github repo just last night— a whopping 900mb btw!
Chez became a point of interest to me after learning about it's solid performance (can even do whole program optimisation and across library boundaries).
I wonder how feasible it would be to build a Chez target for Clojure?
Most recently it has been open sourced under a GNU license by cisco, and is even on Github.
A few years back the Graal compiler from Oracle caught my attention as it was able to accelerate ruby code on the JVM with incredible results.
However it was not able to support any real world code / frameworks— ie. Rails.
(At the time I even downloaded the compiler myself, but it was quite experimental)