Video is by far the largest bandwidth hog of any media type or service on the Internet. According to Cisco’s latest Visual Networking Index, video streaming services accounted for 66 percent of global Internet traffic in 2013. This figure is expected to grow to 79 percent by 2018, an amount of traffic equivalent to six billion DVDs per month. The increase in video traffic is putting a heavy burden on the current Internet infrastructure, causing quality issues for consumers, and spurring a dispute between ISPs and content providers over who is to blame for these issues. Disputes such as the Verizon and Netflix battle created headlines in the past few months, with each side blaming the other for the consumers’ troubles.
What Is the Solution?
Reducing bit rates is one possible solution, but since video quality is directly related to the bit rate allocated to the video stream, blindly lowering the bit rate will result in a poor viewing experience and unsatisfied customers—an option that is unacceptable in the age of high-resolution displays and UHD televisions.
Adaptive bit rate streaming (ABR) is another solution commonly used by content providers. With ABR, the bit rate of the streamed video is selected from a set of available bit rates based on the user’s current bandwidth. While this strategy can provide consistent streaming on high-end and low-end connections, it incurs additional storage and encoding costs, and has a challenge to maintain overall quality on a global scale.
An emerging solution that aims to benefit content providers and users alike is media optimization, the process of analyzing the perceptual properties of an already-compressed video stream and encoding it to a lower bit rate without affecting the original video quality. This would be like taking a ball of modeling clay and squeezing it to make it smaller—it still has the same amount of clay but occupies a smaller amount of space.
How media optimization fits in the workflow.
Media optimization that is content-adaptive—meaning it is applied differently to each video clip based on content, resolution and quality—can reduce the bit rate of each video stream without noticeable visual changes. When done correctly using a reliable perceptual quality measure, this process can reduce bit rate and file size by 20 to 50 percent while retaining the full perceptual quality of the source video file. Media optimization can also be applied on top of existing ABR and caching measures to create the smoothest streaming possible.
In a typical postproduction workflow, media optimization is applied after the initial encoding of the video files but before they are encrypted for secure delivery and packaged for streaming. The quality of each encoded video stream (or ABR layer) serves as a reference for the media optimization process, which retains the input quality while reducing the output bit rate. After the encoded files have been optimized, they can then be encrypted and packaged as usual.
Whether distributors choose media optimization, ABR, bit rate reduction or a combination of the three, the goal is to solve the bandwidth bottleneck problem while providing a win-win-win situation for content providers, telcos and end users. With media optimization, content providers can save a significant amount of money in CDN and storage costs; telcos and distribution networks can benefit from reduced network congestion; and end users can win with smoother streaming experiences, faster stream starts and fewer buffering events. ve
Dror Gill is CTO of Beamr Video, an imaging technology company that is improving the user experience and reducing costs associated with storing and transmitting media files. For more information, visit www.beamrvideo.com.