OpenCL potentials: Watermarked media for content-protection

HTML5 has the future, now Flash and Silverlight are abandoning the market to make the way free for HTML5-video. There is one big problem and that is that it is hard to protect the content – before you know the movie is on the free market. DRM is only a temporary solution and many times ends in user-frustration who just want to see the movie wherever they want.

If you look at e-books, you see a much better way to make sure PDFs don’t get all over the web: personalizing. With images and videos this could be done too. The example here at the right has a very obvious, clearly visible watermark (source), but there are many methods which are not easy to see – and thus easier to miss by people who want to have needs to clean the file. It therefore has a clear advantage over DRM, where it is obvious what has to be removed. Watermarks give the buyers freedom of use. The only disadvantage is that personalised video’s ownership cannot be transferred.

Personalised & Watermarked

Awareness is the most important thing when it comes to copy-protection. People have to know it is watermarked. Consumers don’t have the feeling they are tricked when some organisation knocks at their door – they would have acted more responsibly when they knew their name was printed on the media. In the previous century we bought the box around the DVD, and the content of the DVD is of everybody – the signature of an actor should be put on the box. This century we don’t get a box anymore and we only know we get sued when sharing the movie with friends. Just looking at the past 20 years I cannot say that people have a good feeling of ownership. I believe that using the visual opportunities as shown below, or even putting a personalised message like “for your 2011 birthday, hugs from auntie”, gives the buyer a better feeling of ownership than the 1 minute long warnings before the movie starts.

This 2007 Master thesis by Martin Zlomek discusses various ways of watermarking and how they can be attacked. It is quite honest which are good and which are not – see chapter “robustness”. Recompression, scaling, cropping, noising, denoising, blurring, sharpening all have ways to deform the watermark but take the movie itself with it. Using several movies and blur out differences keeps being a good attack. This means that it is as bad as DRM, once professionals focus on the movie. Difference is that you can never tell where the watermark can be hidden too and thus must make a gamble all is removed when releasing the movie.

When streaming the movie, the renting-company could use the technique to build up a better customer-relation. A personalised advertisement can be added anywhere inside the movie. This could make it possible to show the movie for free, which disrupts the illegal market.

Only disadvantage is that all above costs a lot of processing-power which could raise the price per video too much.

The fast lane

Using video-cards and GPGPU, a video can be watermarked with GPGPU much faster than with traditional techniques. An implementation using GPUs I found here [PDF], but many more possibilities exist. Compression can be prepared quite far (motion-detection, actual compression, etc), before additional data is added. Finishing the video-compression using OpenCL could start delivering the movie for streaming fast.

StreamHPC thinks along with you to get ideas for your problem and gets your product developed. Look around at the site to see what we also do – ask us any question you have via the contact-form.