This is bullshit right?
You can't insert unique ID's into already rendered video as it's being downloaded from the porn site, right? I paid for this and want to share it.
Digital Video Fingerprinting can tell you what a file is, but it's not inside the file. Watermarks can't be added on the fly, right?
Other urls found in this thread:
en.wikipedia.org
sciencedirect.com
twitter.com
wrong
Of course they can.
better start looking for a good lawyer, OP
>You can't insert unique ID's into already rendered video as it's being downloaded from the porn site, right?
There is always a way to embed information into any image, sound, video. You can erase it only with cutting quality.
>You can't insert unique ID's into already rendered video as it's being downloaded from the porn site, right?
Also, it's perfectly possible to make it computationally cheap enough just to embed it on the fly.
>Pay for it twice
>Diff the av stream
>"blur" wherever its different
Oh boy I bet you didn't know your printer put its serial number on every page you print too
> not buying thrice and outputting only whatever content is equal in at least two of the versions.
you do realise you could do the same thing with two copies? dumb american
There are hundreds of patents on that exact thing.
Notably, AACS 2.x Annex B does it in realtime on UltraHD BluRays.
Congratulations, you have removed 1 bit of watermark.
This is true only if only one bit of watermark information differs.
I always ignore this fbi message its none of my biz I just want to fap the fbi can shut the fuck up
with only two copies, how do you have any degree of confidence in which side of diff is the original and which is watermark?
Wouldn't it be hard for them to implement this, because for each client they would have to have a different video file to be able to link it to them?
Does anyone know how this work?
that isn't at all the same thing as inserting a unique id into a file as it's being downloaded so you can know who shared it.
Very easy to do. All they have to do is take the real content and prepend it or append it with a header or footer frame that's uniquely coded to you. Anybody programmer familiar with video encoding could make a basic version of it in a day.
Wasn't this how they caught those people who leaked those Oscars screeners some years ago?
You serious? How?
They print a series of small yellow dots that encode the printer serial number and a timestamp onto everything printed.
You only need about 128bits of identifyable information for any piece of work. Photos and Music can be altered when transmitted to provide an unique ID. single bit color changes and pitch adjustments troughout the file wont be noticed by anyone. You need to keep the original file and algorithm hidden. Else it can be scrambled.
stallman was right, even printers are botnet
They had audio and/or visual watermarks
They scan the frames not look for watermarks.
>familiar with video encoding
You just need to be familiar with i/o in binary files and know a few details about the file you are editing.
Digital signatures or anything else could be inserted in the file's specification comment or additional info field.
It doesn't have to be an extra frame.
It's not only not bullshit, it's happening, and it's only going to get worse. Should check out what PrimeLeap's doing with image DRM. :^)
just run it through davinci resolve and re-encode it
takes 5 minutes at most and it will destroy any watermarking
Can't you just like.
Flip it horizontally and reencode it in an anti rely different code or something
Sorry for my phone posting plead spit on me
could mean any number of things, such as;
a. nothing, just a scare tactic
b. unique file metadata (trivial to add, trivial to remove)
c. partially/fully watermarked video (can be expensive to add for each download, but definitely possible)
d. partially/fully watermarked audio (not as expensive)
given enough cpu time, anything can be added in realtime
there are video/audio watermarking technologies that can withstand lossy transcoding
>a
Makes the most sense
>b
Possible
>c
No fucking way they would do that for mainstream media distribution.
>d
Basically the same as c.
Just move to Sweden
>No fucking way they would do that for mainstream media distribution.
by 'partial' i mean something like;
1. uniquely watermark/encode 5 second warning message
2. stream-concatenate that with the main video
3. serve that
it's expensive relative to nothing/metadata-only, but still pretty feasible
There was that one streaming site who would watermark release by appending a signature at the end of the file.
Obviously that didn't go over so well for them once someone started dropping releases of their stuff with randomized sigs, they went with some other watermarking and I know they changed method since but I dunno how they did it.
>I paid for this
you should never pay for porn
They cant possibly be talking about a hidden watermark. When the video is re-encoded, its slightly blurred and all hidden (non visual, data only) watermarks would get lost
>he doesn't know about fault tollerant watermarking
clueless retard
>Color cartridge is busted dry and empty for years.
CIA niggers BTFO
>he doesnt provide any real-world proof
paranoid fuck
sciencedirect.com
and that's just the tip, next time do your own research before you look like a baby.
>No fucking way they would do that for mainstream media distribution.
Not that hard if you only do it for a few certain parts of the video.
From your own source
>6.1.2. Remodulation attacks
>Since lossy compression and denoising have been widely presented in the literature, with some applications of low bit rate coding and image enhancement, respectively; it is not incredible that they are also famous attack tools for the watermarking community. On the other hand, remodulation attacks are a rather fresh theory unique to the watermarking attacks. A systematic remodulation attack was first demonstrated in [32]. In this algorithm, the watermark was forecasted using subtracting from the host stego image to the median filtered version of stego image. The forecasted watermark was also truncated, high-pass filtered, and the subtraction is done from the stego image with a constant amplification parameter with2. Since the median filtering mainly takes away the noise in the high-frequency section, the low-frequency section cannot correctly estimate the value according to this filter. In the situation of a highly consonant between the amplification parameter and the estimated watermark, the attacks have the guidance to a diminishment in extensive correlation in the matched filter with decoding.
get wrecked, son
>and that's just the tip
Go and read some more, dumb kid.
>hit up server for video
>video is rendered in three parts
>first and last parts are 99% of the entire content
>1% of the content is dynamically fucked with to include your info
>video is rendered
Do-able.
If you paid through some sort of account or with your name they might have embedded a watermark associated with your name or username.