The Future of VR - Ep 3: Interview with Forest Key of Pixvana

An equerectangular still from the 360 VR Video "Zapallar Chile" from director Carlos Key, currently in production after the Key family's spring break visit to Chile.

An equerectangular still from the 360 VR Video "Zapallar Chile" from director Carlos Key, currently in production after the Key family's spring break visit to Chile.

While video games have dominated virtual reality content, 360 video is on the rise and is a key element in stoking popularity among early adopters and VR enthusiasts.  Live music, sporting events and travel content have lead the way, with companies like NextVR, Jaunt and LittleStar raising capital and making deals with the NBA, Conde Nast and ABC News to deliver VR experiences.  

While watching 360 video in VR headsets is compelling, often the quality of the video is still very low and hard to watch.  A lot of that has to do with the way the files are created, compressed and delivered.  Pixvana, a Seattle VR startup, founded by Forest Key, has some of the answers to these problem and is one of the early pioneers in 360 video. Pixvana has created a new cloud-based platform for creating and delivering ultra high-resolution VR video, empowering content creators and businesses who use VR for communication.  

I recently had the chance to ask Forest what his thoughts were about the future of VR and the interesting trends he sees on the horizon.  Below is a summary of of his perspectives.


Forest Key | Pixvana

Forest Key is founder CEO of Pixvana, a Seattle based VR startup that is focussed on video processing and delivery to "XR" headsets/applications as a cloud infrastructure service layer. Forest was previously the founder and CEO of buuteeq inc., a SaaS marketing automation provider to hotels, which was acquired by Priceline and now operates as BookingSuite, a division of Booking.com.  As a product manager he worked on both Microsoft Silverlight and Adobe Flash and has been involved in web video core technology since the inception of the medium.  He started his career in the visual effects / CGI space as a CGI artist while working at Lucasfilm.  Forest studied History and Film at UCLA and is a alum of Palo Alto High School.


Q1: When will VR AR MR go 'mainstream'? How long will it take to get there?

The verbose answer is that when i think about the vr/ar/mr technologies going mainstream i think of "XR"--as in "x-reality".   VR and AR only exist as separate concepts today because the headsets are quite distinct and in their prototype phases.  100m consumers are not going to buy these headsets, rather, a device that is a few years out that we haven't seen yet but that we in the field clearly anticipate.  This "XR" headset will be able to switch between VR (opaque) and AR (transparent) modes that either replace or augment the "reality" of the world around us.  The XR headset will be lighter, cheaper, and better than anything we have on the market today, and be akin to the first "iPhone" like product that gains mass adoption.  Everything between now and then is just dress-rehearsal, where software companies like Pixvana are building out our underlining platforms and tools to get ready for these XR devices.  The short answer -- 5 years.

Q2: What currently are the most interesting trends in the VR AR MR (XR) space?

I think there are 3 primary trends in XR: the innovation in hardware, software, and content.

In hardware we are seeing a rapid improvement in the power/weight/cost of head-mounted-displays for both AR and VR, and i think we can expect a 2x better, 50% cheaper trends for several years.  For example--in 2018 we will have sub $400 6-degrees-of-freedom headsets on the market, down from 2x that price a year ago.  This is going to be driven largely by Microsoft, who are introducing a bevy of features in Windows that will make building a great VR/AR headset that runs on Windows much more cost-effective and high quality.

In software everything is in play because almost nothing is currently available.  Pixvana is building an XR Video production and publishing pipeline--think of Quicktime in mid 1990s.  Others are building social interaction layers, communication stacks, game engines... all the pieces necessary for the eventual build out of the killer apps in the ecosystem.

And in content creative teams are just starting to experiment with what makes a killer experience in the medium.  Just like when the film camera was invented in the 1890s, the first experiments have more to do with what we know from our prior experience with games and movies, and less to do with what actually will be the killer content in this new medium.  It will take several years of iteration to nail the unique vernacular and syntax for how to tell a great story in XR and create XR specific user experiences that are compelling to consumers.

Q3: How do you see XR Video capture and distribution evolving and what is it that most companies are missing?

XR Video depends on innovations in the cameras, the software tools to create stories, and in the creative advancements in the stories themselves. 

Today the cameras are very primitive and mostly are hacks of existing cameras into rigs/arrays that allow 360 degree coverage.  A great example of these is the GoPro based cameras, that actually do a pretty darn good job and can capture a 8k 360 degree video if you know how to use them.  Facebook, Google, and Jaunt have built cameras that use as many as 16 separate lenses and sensors and then try to extract stereo depth maps to create 3d video--these generally are very complex and only work some of the time.  In ~5 years we'll have phone based 360 capture cameras in our pockets, that will not only capture the Red Green and Blue light information from the scene, but also the depth and volume information--what is called "volumetric" or "light-field" video, which has the added benefit of depth and realism that really makes for "oh my god that's cool" VR Video experiences.  Before then, we'll likely see very good, purpose designed cameras from the likes of Sony, Canon, and Nikon--traditional camera manufacturers.

Software tools for creating and publishing VR Video need to be re-thought for XR specific mediums.  XR Video needs to be 10k or higher resolution so that the effective resolution of the video in your field-of-view in the headset is high enough resolution to fill the headsets screen--this is a requirement that really pushes legacy production workflows that have been focussed on 2k and sometimes 4k.  A 10k resolution video pipeline entails 10x the processing and thus really slows down the creative experimentation.  Pixvana is building all of these production tools as cloud systems, where we can spin-up 100 processors to render a job, and thus bring the interactivity back into the process so that a storyteller can experiment and quickly try different ways to edit a story and get it down to the headset for viewing at high quality.

And when it comes to content, we are already seeing great VR Videos being made by the likes of traditional media companies like the New York Times (who are making a 360 video every day of the week, usually on documentary/news subjects) and native VR only media companies like Within or WeVR (who are creating VR native stories and series).  What is uniquely compelling about the XR medium is the sense of empathy and presence that you can give the viewer--sports, concerts, celebrity encounters, and documentary films are some of the early use cases that are undoubtedly compelling--but more time and experimentation is needed.

Forest's teenage son, director Carlos Key, shooting a documentary VR film about the town of Zapallar Chile using a VR video camera, the GoPro Omni

Forest's teenage son, director Carlos Key, shooting a documentary VR film about the town of Zapallar Chile using a VR video camera, the GoPro Omni


Key Take Aways

  • VR/AR/MR technologies will converge into an xR Headset that will be able to switch between VR (opaque) and AR (transparent) modes.  Once this happens we will start to see mainstream consumer adoption.
  • We can expect to see lighter, cheaper and smaller VR hardware as software companies and content providers try to figure out how to deliver compelling experiences and set the stage for the next killer VR app.
  • Advances in XR Video will depend primarily on innovations in the cameras, the software tools to create stories, and in the creative advancements in the story-telling.
     

Huge thanks to Forest Key for agreeing to contribute his thoughts and ideas to this interview.


Boaz Ashkenazy is Co-founder of @Studio216.  Studio216 is a immersive technology agency focused on VR/AR/MR for the enterprise.   He is also a dynamic entrepreneur, speaker and writer from Seattle who is examining and envisioning the future of immersive technologies. Boaz can be reached at boaz@studio216.com or on twitter @boazashkenazy