Vidyo recently celebrated its 10-year anniversary. It’s amazing to contrast the range of products that we offer today, and the number of people that we reach every day through our software solutions with the first meetings of the early days. What started off as mere concepts in whiteboards and presentation slides has translated to technology that has changed our industry in very significant ways.
Vidyo’s software based approach limits the need for transcoding and associated infrastructure.
I could describe these changes in various terms – changes in product architectures, standards, industry best practices, and so on – but there is a great measurement of innovation that has the benefit of originating from objective third parties: patent offices. As it turns out, Vidyo recently celebrated the issuance of its 100th worldwide patent.
Vidyo was founded in order to build a disruptive, new architecture for large-scale, very high quality video communication, on virtually any network and device. From day one, as a new company in a space of considerable technological complexity, it sought to protect its innovations through patent applications filed in the US and elsewhere around the globe. These patent applications started producing results within a few years, with our first patent being issued in the US on September 22, 2009 (US Pat. Nr. 7,593,032) and the 100th being issued in the US on July 7, 2015 (US Pat. Nr. 9,078,004). As of today Vidyo has a total of 106 patents worldwide.
Vidyo’s patents span a wide range of subjects. Some relate directly to our products, some have been found essential to practice certain industry standards in video compression and communication, and some describe more strategic technology explorations. Several apply to more than one category.
Although each patent has its own history and background, I chose to highlight here two particular patent “families.” (By patent family we typically refer to an invention for which patents are being pursued at multiple jurisdictions around the world, and which may result even in multiple actual patents in the same jurisdiction.)
The first family happens to include our very first issued patent, US Pat. Nr. 7,593,032. The title, “System and Method for a Conference Server Architecture for Low Delay and Distributed Conferencing Applications,” describes certain aspects of the operation of our VidyoRouter. The VidyoRouter is a cornerstone of Vidyo’s technical architecture, in that it takes advantage of scalable video coding to deliver an unprecedented range of features (error resilience and localization, rate adaptation, personalized layout, cascading, etc.) with extremely low delay (an order of magnitude less than a typical MCU) and at a massive scale – and all that without any signal processing at the server. The VidyoRouter’s design did not exist as a concept in the Internet Engineering Task Force (IETF) specifications when it was first introduced in our products in 2008. For example, the RTP Topologies RFC, RFC 5117, published at the same time, did not foresee it. We have subsequently introduced the concept of the “Selective Forwarding Unit” (SFU), now incorporated into the RTP Topologies update (which should replace RFC 5117 soon), to cover this design using a generic, non-commercial term.
The second patent family that I would like to highlight includes US Pat. Nr. 9,077,964, “Systems and Methods for Error Resilience and Random Access in Video Communication Systems.” As the title suggests this patent deals with error resilience. It covers a key concept for achieving error resilience through the use of temporal scalability. The concept is very simple: by using temporal scalability, one merely has to protect the lowest temporal layer, typically one fourth or one half of the original pictures, rather than the entire set of pictures. Since these pictures are now at two or four times the temporal distance, the system can use retransmission to recover any parts of them that may be lost. Using a design that allows the system to detect such loss with the very first packet that it correctly receives, one can achieve error resilience that is an order of magnitude higher than techniques that use non-scalable video. At Vidyo we decided that this technique would be of significant value to the entire industry and so proposed its inclusion in all RTP payload formats of relevant codecs: H.264 SVC (RFC 6190), H.265, VP8, and VP9. We submitted intellectual property rights (IPR) disclosures at the IETF that ensure that people who implement these standards can use it*. At the same time, certain claims of the 9,077,964 patent and other Vidyo patents have also been found essential for implementing the H.264 SVC and H.265 codecs, and we have licensed the relevant claims through the corresponding patent pools managed by MPEG-LA, thus making it easier for people to make use of these standards as well.
These two patent families highlight two important dimensions of Vidyo’s technology development efforts, which form a continuous balancing act. Some innovations we maintain exclusively to Vidyo, as they are part of what makes Vidyo unique. Some others we believe are better suited to be incorporated into standards so that anybody can have access to them, thus benefitting the entire industry – both producers and users (ourselves included, of course!). As the industry matures further, one can expect more and more technical features to become standardized, as new ones come in to define differentiation. This has been the traditional development cycle of standardization and innovation for a long time, and one in which we expect to continue to be an active participant.
It’s been a fascinating journey – and a real privilege to be part of the team that is making it happen. On to the next 100!