For those of you actually reading this thread - it means you are actually interested in piracy prevention and proper detection....

The "real world" is light years ahead on all this as visual hashing is a massive research area (especially in videos given transcoding, youtube, copyright etc)

For one example (warning scientific paper, but well written and good visuals!) - see this PDF manuscript. There are hundreds more like it (but more mathematically detailed!)

The real world is moving forward a lot, which is why large companies can easily send take down notices to google because things are automated.

When it's so easy for some to send out mass emails to end-users of pirated content, why the need for all this piracy prevention?

I really do find it extremely amazing that there has been only one person to date that has actually contacted me to discuss wanting to implement things discussed in this thread on their sites. (4 days on the front page is like 8 years in real human terms)

If piracy is such a major problem in this industry, I only see a handful of people interacting in a serious thread that is highlighting how to combat it :Oh crap

Timely bump for people to digitally fingerprint their own content easily...

Bump for borked's solution.

Now this is getting interesting. I didn't like the branding each stream with a username and IP as I thought it was an ugly idea, and branding each customer as a thief. In stores, they remove the security tag on purchase.

However, this hashed pixel idea is brilliant.

I'm a magician as a hobby, and have several magic content makers that would love this technology/idea Borked.

I assume you are OK with mainstream work and I can tell them about your ideas?

Thanks

Damian

Branding videos with username IP was just an idea, but yes visual hashing is where the mass screening detection lies.

I have no idea what you have as an idea, but go ahead!

Nominated for Thread of the Year... great business thread.

UPDATE - compare images on the fly


OK, I got very intrigued by all these algorithms, so much so, I spent many many sleepless hours these last few days trying to get my head around them (already 1am here!)

I've got something I think is pretty darn smart in comparing images here...

Please give it a whirl with your own images and let me know what you think....

Try to make them as different as possible....

http://borkedcoder.com/image_comparison/

Report back here your ideas + suggestions....

tested with...

http://borkedcoder.com/image_compari...d9357935f3.jpg

http://borkedcoder.com/image_compari...6cf9aac735.jpg

gave:

The two images below are very similar and would be flagged as pirated!

This would work for my sites. I would love to have software that can do this.

If I could figure out which member was the downloading maggot, I could have them permanently banned from re-joining my site or any other ccBill provided site.

help please...
 

I am actively looking for a way of protecting the content on my sites. I am pretty attracted to the idea of a non-downloadable, anti-leech streaming system like the way Adobe and some other companies are boasting.

I called my billing company (ccBill) and they are promoting metrixstream. I spoke to them and their monthly licensing fee is pretty steep.

I am open to all suggestions that will slow down these file sharing maggots who rip off material for a measly membership fee and post my entire site on 10 different forums.

I really don't want to piss off my members, but more importantly, I just want a big deterrent to file sharing.

Any suggestions would be appreciated, and I am open to spending money on this, if it's a good service.

Hoping for some good help!

Great thread borked! this could be useful for many adult webmasters!

Wish someone could inplent a pincode into a movie or something verifying the user as a "member"

Awesome thread.

If you read further into teh thread, there is a way to do this on the fly, without the overhead of re-encoding.

You can produce a say 5 second roll at the start of the movie that says, "this movie was prepared for <username> <IP> on <data>. Copyright <your company> All rights protected. Do not pirate this movie.

That would scare a big bunch away from uploading it. For those that know how to remove frames, you would also inject that into the movie "randomly", injected as a single frame and the end user will never see it.

It's not the same as digital fingerprinting, but it's a good way of tagging the content.


I would suggest going with http://www.wowzamedia.com/store.htmlWowzaMediaServer. At $65/mo for a license, it will server all your needs. If you need help setting it up, hardening it, and optimising it for streamlined production environment, drop me an email.

you mean like the injection of user details/identification data as in post 76?

the data in the example is in a single frame at exactly half way into the movie.

I wasted lots of money, with some fancy programming firm in Dallas, then finally said enough. This information is great! Thanks!

an example of identifying the user in both preroll and injected somewhere in the movie

download
http://borkedcoder.com/movie_injecto...er=borkedcoder

the identifying data (given as example) is injected at the start of the movie and at exactly mid-point (search frame by frame at 8 secs, just where the boy starts walking) and at the end

Nice writeup

That's clever :thumbsup

I've just lost 1¼ hours thinking about it :upsidedow

The two images below are very similar and would be flagged as PIRATED!

http://media.ferrocash.com/photo/201...o-test-001.jpg

http://media.ferrocash.com/photo/201...o-test-002.jpg

Real cool job pal! :thumbsup

d7f3bf4563.jpg

:)

Close as I can get at the moment and get NO PIRATE.

oooh, you are sneaky!

OK, took care of that kind of detection avoidance by tightening up the crop-detection part....

cropped tight and a spherise filter applied...

The two images below are very similar and would be flagged as PIRATED! :thumbsup

http://borkedcoder.com/image_compari...9ef4150dd4.jpg
http://borkedcoder.com/image_compari...2af49c5c9b.jpg

thanks for catching that :thumbsup

anyone with any false positives? something I'm worried about is a very similar scene close up but from 2 very different movies...

Cropped, resized, distorted, severely recompressed, rewatermarked (typical tube action). And still:

The two images below are very similar and would be flagged as PIRATED!

http://media.ferrocash.com/photo/201...o-test-004.jpg

http://media.ferrocash.com/photo/201...o-test-005.jpg

Pretty cool thingy.

2 borked

How many images can your script compare in, say, 1 hour?

Damn...

The two images below are very similar and would be flagged as PIRATED!

http://media.ferrocash.com/photo/201...o-test-004.jpg

http://media.ferrocash.com/photo/201...o-test-006.jpg

thanks - that's what I wanted to see why a false positive came up... will look into it

btw, I've added a timer to the image comparing (doesn't include upload speed)
eg
time taken to compare images: 2.46392607689 seconds (~1461 images/hr)

I've added the confidence level whenever the images are flagged as pirated.

It was set to anything > 70%, flag as pirated - your images were 72%.

75% is very good at not letting through false positives through, but with the problem of letting some false negatives through.

It is currently set to 73% or greater to be flagged.

Would still get flagged even by raising the bar as the images are 76% similar....

The more "noise" in the image, the better.

A big full-frame close up of a pussy and well, one pussy looks like another, so false positive rate is higher...

Too slow unfortunately. To make it practical, it should do at least a billion comparisons/hour.

Nearly everything I have tried doing has been with an aim of keeping the image as close to the original as possible and get it to pass as NO PIRATE.

Just think that to make your comparison test as good as possible I need to to be trying to think of a way I could get a whole photoset to pass while still having it acceptable to the human eye.

Screen capturing video, at least in my experience with using Camtasia, loses a bit of quality (especially in audio, and the videos jump a bit depending on the PC you're using to capture it with). The other aspect is that the user has to actually sit there for the entire feed to rip it -- for a lot of people, that's too much work....and not worth it.

well, to be fair, this is running on a dev server that is about 5 years old. Shit, it takes me about 15 minutes to compile ffmpeg whereas on a new quad core, it's compiled in ~15 secs.

Plus, there could be many instances of the app running - no need for a single fork to be doing all the work...

And of course, you would only search for comparisons after searching for pages that flag your keywords... you wouldn't have to search ALL photos on ALL sites

--edit

ok, I see what you mean - pairwise comparison of a single photo with all your photos.

Can you send me a zip with some photos in (more the merrier, but try to keep the zip <5MB) and a test image to compare against? The script will run *much* faster comparing 1 photo to many than doing 1on1 comparisons...

Similar to what I said above, you could lower the level to whatever is satisfactory to *not* let false negatives through, which would allow more false positives to show up. However, I suppose this stuff is with the view to scan loads of images per hour and to flag ones that require further intervention by a human eye.

Automated scripts will always cause false negatives and positives, but false negatives are much more costly. Much better to let a human brain search through the results flagged as pirated and toss away the false positives, than it is to allow false negatives slip through.

But yes, you're right - if a photoset contains 15 images - it only takes 1 to be flagged as pirated to raise the alarm bell on the photoset.

Is is possible to estimate it's maximum productivity on a good modern server, and take into the account both server and script optimization (multi threaded mode or some other tricks)?

I understand that you need to optimize where you search and what finds you're going to compare, but still - we have a database of about one million pictures to protect, camparing against this db by 1Kpics/hour is kinda not going to work. Even 100K/hour not going to work.

A good post and an excellent read. Thanks.

The only problem with it is the part where you wrote "The only way to "rip" your stream is to have a screen capture program record full playback of your move. Impossible to prevent that!" That's a very weak Achilles heel to the whole 'solution.' It means essentially, YES your videos can still be pirated. And once *anyone* does it, they have a pirated version of the video which they can use to propagate tubes, torrents and other p2p media exactly the same way they would have if they had downloaded it. The basic problem is that you only need 1 competent thief and all the incompetent ones get access to whatever he stole exactly the same way as if they stole it. :2 cents:

This allows you the opportunity to upsell and try to re-convert them as welll.

this is exactly what has been implemented - great return rates on those coming back and trying to view a movie they are not entitled to view...

Blocking users from logging in simply because their pass expired is dumb... give them the members area but on trying to view something that they are not entitled to view... upsell, like you said :thumbsup

Yes, I think I have a way around that... pre-calculate all your images before hand and dump the floating hash. That way, the app simply has to calculate from the single image and compare it for likeness to all the pre-calculated ones.

Let me see if that's possible - that should speed shit up *enormously* (well, except for inital calculations on you image stock). I will put that on "to do" for next week as I'm a bit loaded up atm due to time I wasted in doing this app :Oh crap

then watermark your streams with the user's username, IP etc - this was discussed in the thread. It's impossible to stop a screen grabber from ripping your streams, but you can deter it perhaps :2 cents:

I just heard the FSC, Mansef and Top Bucks just put a hit out on you. :upsidedow


You know I'm waiting patiently. :thumbsup

Just sent you mail BTW about your dog pic.

How do you implement this with 3rd party billing? At least for an upgrade back into the site after they cancel, I don't think that is possible unless you have their date and your own merchant account, correct?

Don't have to go into details here, can shoot me an email.