When we started FutuLabs here at Futurice, the idea behind it was that if we want to do new and exciting projects, we need to know our way around new and exciting technologies. Our clients trust our technological expertise, and for that we need to get our hands on the latest gadgets as soon as possible.

The solution was to just do it and thus FutuLabs was born.

Like most things we do here, we started out lean, got our hands on some of the most exciting gadgets we could think of (that could potentially be used in projects in the future) and started playing around. It seemed like a fantasy that we would actually get paid to work with these things, but that is precisely what happened.

A future panorama

Early this summer we had an opportunity to talk to a Finnish financial institution about a new space they were planning. The space would have a strong focus on new technology, so it was a perfect place to explore what the future of real estate and financial services could be like. Our clients welcomed the chance to come to our office to see and hear our ideas.

We prepared some demos using the Oculus Rift virtual reality glasses. The demos were focused specifically on building and real estate applications, but we couldn’t pass up the chance to show them a roller coaster. The demos were a huge success and thus began the first project to come out of the FutuLabs initiative.

After investigating different alternatives, we decided that we would film a spherical panorama of a number of apartments and show these in this new space using the Oculus Rift. First order of business: Buy 10 GoPro cameras and a 3D printed array to hold the cameras pointing in every direction.

Unexplored territory

Usually we start a project by setting up the build environments and CI machines. Not this time. The first weeks of the project were full of unpacking and building. Filming, configuring, video editing and more filming. Turns out stitching together ten video streams into one perfect spherical video is not a trivial task and there are no npm packages for this (yet).

As we had the privilege of hosting the June HelsinkiJS event here at our Helsinki office that became one of our first test videos. One battery was DOA and one camera was set to the wrong mode which meant that a small portion of the space is not in the video. This is why we test early and iterate. Still, not a bad first try. Check out the video here: https://www.youtube.com/watch?v=bhRg4Woc9Fs

At this point we really started to see how much both active and passive time the video stitching and editing really takes. Our clients were happy with the results we produced, but we’re anxiously anticipating the next generation of 360 cameras that seem to be just around the corner. These new cameras will (hopefully) eliminate the stitching part of the process which is currently the main stumbling block for great videos. You rarely see spherical panorama videos that do not have some errors where the different video streams meet.

Finally, we needed the software to show the videos we made. There are no software products on the market that would give us everything we needed, and besides, custom software is what we do. Unity was chosen as the technology platform to build this on. It has some good support for the Oculus and allows for rapid prototyping and development.

Testing the controller for the viewer

The end result of all of this was one of the strangest and most exciting projects I’ve been a part of as well as a happy customer. Working with new technology is definitely challenging, and this still has to be considered more as a proof of concept, but who knows what the future holds? One thing is for sure, FutuLabs will continue and it will enable us to stay on the cutting edge where we thrive.