3D STREAMING TECHNOLOGY AND LIMITATIONS
3D films are not a new technology, but only recently have movies been brought into our homes thanks to the high capacity of Blu-Ray discs and because of new televisions that feature high refresh rates which show films at double the frame rate required for our eyes to distinguish motion. A 3D movie is really two entire movies shown at the same time, one that your left eye sees, and one for your right eye. Special glasses are required to switch between the two in a way that your brain perceives as three-dimensional. Because of this, twice as much data is needed for a 3D movie compared to a 2D one, which presents certain issues when trying to send this data over a medium such as the Internet. Thankfully, several technologies are emerging that will make this possible in the near future and that we believe Blockbuster is well positioned to take advantage of in order to offer 3D content streaming to their customers.
The biggest factor of sending a movie over the Internet is the amount and type of compression that is used. It would be impossible to send a fully uncompressed high definition movie directly to your home, whether or not it is 3D, because the amount of data is staggering. In order to make the files smaller and able to be sent in small pieces, companies like Netflix use compression formats such as MPEG or AVI which use special technologies that result in smaller files. This involves limiting the bit depth for colors in a way that humans don’t really perceive, as well as comparing individual frames to the ones ahead of and before it so that only the differences have to be shown. In an MPEG-4 video, the type used by Netflix and Apple, scenes with very little action such as in a dark room use much less data compared to an action scene because of the way this compression scheme works (IEEE, 2004).
On a Blu-Ray disc, they simply splice both video channels into a single stream with twice the data rate, alternating frames in a single video file so it is easier for the television to output the video. All it has to do is play twice as many frames and it simple works, because the video files are not encoded in any way.
For streaming in 3D, encoding will be necessary but the technique of alternating frames isn’t ideal because the difference between individual frames is so high it would appear similar to the worst-case action movie in the previous example. One technique being worked on by researchers is to calculate the difference not only between adjacent frames, but also between left eye and right eye channels, because they are likely to be similar even though they are not exactly the same (Zamarin, 2010).
In an mp3 music file, for example, the data that is encoded is not both the left and right channels, but instead one channel and another set of data that represents the difference between the left and right channel (Wilburn, 2007). This takes much less data and can be easily decoded. So far the difficulty has been in finding an algorithm that can detect these differences between stereo images, and combining them with the existing MPEG algorithms to allow streaming over the web with an acceptable amount of lossiness. One standard already exists that supports stereo video and is an expansion of the MPEG-4 codec called MPEG-MVC, which stands for “Multiview Video Coding.” However, this spec is not intended for direct streaming use, and may not be the most optimal for this application (Smolic, A. 2008). Another consideration is that the lossiness, or the amount of video degradation, also needs to be variable because different viewers have different qualities of connection and the quality of a viewer’s connection may change over time during the viewing. Compression technologies exist for this as well, such as Z-lib which reduces the number of colors available and which works well for video because colors, such as people’s skin, doesn’t change rapidly. (Yang, 2006).
Through a combination of these compression technologies, it is very conceivable that streaming of 3D movies will become mainstream in the very near future. The largest amount of work will be computing and encoding the videos into these formats, which requires that a standard format be selected, and that film studios sign on for the rights to make their films available in such a format. Because of Blockbuster’s position as a successful streaming provider through it’s subscription and Dish Network services, we believe they are the best candidate to negotiate these contracts and begin delivering the technologies that will allow their customers to stream high-definition 3D video in their own homes.

References (Kris Simonsen)

Smolic, A. (2008). Introduction to Multivie Video Coding. International Organization for Standardization. Retrieved from : http://mpeg.chiariglione.org/technologies/mpeg-4/mp04-mvc/index.htm
Zamarin & Forchhammer (2010). 3D Video Compression and Transmission.
Retrieved from: http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=0CCwQFjAB&url=http%3A%2F%2Fpnrl.stanford.edu%2Fworkshop_2011%2FSession%25203%2FDay1_2_Zamarin.pdf
Yang, Cui, Anwar, Boccino, Kiyanclar. (2006). Real-Time 3D Video Compression for Tele-Immersive Environments.
Retrieved from: http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CEAQFjAA&url=http%3A%2F%2Fwww.cs.cmu.edu%2F~rbocchin%2FPublications_files%2FYang-MMCN-2006.pdf
Wilburn, Thomas (2007). The AudioFile: Understanding MP3 Compression.
Ars Technica. Retrieved from : http://arstechnica.com/old/content/2007/10/the-audiofile-understanding-mp3-compression.ars