Just in case anyone secretly had the idea that Valve Software’s VR and other hardware somehow sprang fully-formed from a lab, here are some great photos and video of early prototypes, and interviews with the people who made them. Some of the hardware is quite raw-looking, some of it is recognizable, and some are from directions that were explored but went nowhere, but it’s all fascinating.
The accompanying video (embedded below) has some great background and stories about the research process, which began with a mandate to explore the concepts of AR and VR and determine what could be done and what was holding things back.
One good peek into this process is the piece of hardware shown to the left. You look into the lens end like a little telescope. It has a projector that beams an image directly into your eye, and it has camera-based tracking that updates that image extremely quickly.
The result is a device that lets you look through a little window into a completely different world. In the video (2:16) one of the developers says “It really taught us just how important tracking was. No matter [how you moved] it was essentially perfect. It was really the first glimpse we had into what could be achieved if you had very low persistence displays, and very good tracking.” That set the direction for the research that followed.
[Daniel Perdomo] and two of his friends have been working on a mechanical version of Pong for the past two years. We can safely say that the final result is beautiful. It’s quite ethereal to watch the pixe–cube move back and forth on the surface.
[Daniel] has worked in computer graphics for advertising for more than 20 years. However, he notes that neither he nor his friends had any experience in mechanics or electronics when they began. Thankfully, the internet (and, presumably, sites like Hackaday) provided them with the information needed.
The pong paddles and and pixel (ball?) sit onto of a glass surface. The moving parts are constrained to the mechanics with magnets. Underneath is a construction not unlike an Etch A Sketch for moving the ball while the paddles are just on a rail with a belt. The whole assembly is made from V-groove extrusion.
Our favorite part of the build is the scroll wheel for moving the paddle back and forth. For a nice smooth movement with some mass behind it, what’s better than a hard-drive platter? They printed out an encoder wheel pattern and glued it to the surface. The electronics are all hand-made. The brains appear to be some of the larger Arduinos. The 8-bit segments, rainbow LEDs, etc were build using strips glued in place with what looks like copper foil tape connecting buses. This is definitely a labor of love.
It really must be seen to be understood. The movement is smooth, and our brains almost want to remove a dimension when watching it. As for the next steps? They are hoping to spin it up into an arcade machine business, and are looking for people with money and experience to help them take it from a one-off prototype to a product. Video after the break.
It’s great to see different kinds of hardware and software tossed into a project together, allowing someone to mix things that don’t normally go together into something new. [Freddy Kilo] did just that with a project he calls his VR Robot Tank. It’s a telepresence device that uses a wireless Xbox controller to drive a tracked platform, which is itself headed by a Raspberry Pi.
The Pi has two cameras on a pan-tilt mount, and those cameras are both aimed and viewed via a Google Cardboard-like setup. A healthy dose of free software glues it together, allowing things like video streaming (with U4VL) and steering via the wireless controller (with xboxdrv). A bit of fiddling was required for some parts – viewing the stereoscopic cameras for example is done by opening and positioning two video windows just right so as to see them through the headset lenses. It doesn’t warp the image to account for the lens distortion in the headset, and the wireless range might be limited, but the end result seems to work well enough.
The tank is driven with the wireless controller while a mobile phone mounted in a headset lets the user see through the cameras; motion sensing in the phone moves those cameras whenever you move your head to look around. Remote Control hobbyists will recognize the project as doing essentially the same job as FPV setups for model aircraft (for example, Drone Racing or even Snow Sleds) but this project uses a completely different hardware and software toolchain. It demonstrates the benefits of having access to open tools to use as virtual “duct tape”, letting people stick different things together to test a concept. It proves almost anything can be made to work if you have a willingness to fiddle!
The web is abuzz with the news that the Facebook-owned Oculus Rift has buried in its terms of service a clause allowing the social media giant access to the “physical movements and dimensions” of its users. This is likely to be used for the purposes of directing advertising to those users and most importantly for the advertisers, measuring the degree of interaction between user and advert. It’s a dream come true for the advertising business, instead of relying on eye-tracking or other engagement studies on limited subsets of users they can take these metrics from their entire user base and hone their offering on an even more targeted basis for peak interaction to maximize their revenue.
Fortunately for us there is a choice even if our community doesn’t circumvent the data-slurping powers of their headsets; a rash of other virtual reality products are in the offing at the moment from Samsung, HTC, and Sony among others, and of course there is Google’s budget offering. Sadly though it is likely that privacy concerns will not touch the non-tech-savvy end-user, so competition alone will not stop the relentless desire from big business to get this close to you. Instead vigilance is the key, to spot such attempts when they make their way into the small print, and to shine a light on them even when the organisations in question would prefer that they remained incognito.
Oculus Rift development kit 2 image: By Ats Kurvet – Own work, CC BY-SA 4.0, via Wikimedia Commons.
Let’s face it: 3-dimensional odometry can be a computationally expensive problem often requiring expensive 3D cameras and optimized algorithms that can be difficult to wrap our head around. Nevertheless, researchers continue to push the bounds of visual odometry forward each year. This past year was no exception, as [Christian], [Matia], and [Davide] have tipped the scale in terms of speed with an algorithm that can track itself in 3D in real time.
In the video (after the break), the landmarks are sparse, the motion to track is relentlessly jagged, but SVO, or Semi-Fast Visual Odometry [PDF warning], keeps tracking its precision with remarkable consistency, making use of “high frequency texture” as a reference. Several other implementations require two cameras or a depth camera variant, but not SVO. It uses a single camera with a high frame rate between 55 and 300 frames per second. Best of all, the trio at the University of Zürich have made their codebase open source and available as a package for ROS.
If you’re looking for a quick and easy project to get into virtual reality, making your own VR skateboard controller is actually pretty easy to do!
First you’ll need some kind of VR headset. You could buy a fancy one, like the Oculus, or a Samsung Gear VR — or you could use something as simple as Google Cardboard — and you could even make your own. All it takes is a phone, an Arduino, a Bluetooth module, and an accelerometer-plus-gyroscope IMU.
I guess if you are going to build a robot to do something boring like telepresence, you might as well make it cute. That’s obviously what [Andrew Maurer] was thinking when he built a telepresence robot using a Wall-E toy. The result is kind of adorable: Wall-E is holding the 5-inch HDMI screen that shows the video, and can scoot around in true Pixar fashion under remote control.
It’s also a neat build on the inside, using a Raspberry Pi for the brains and an Adafruit MotorHat to control the motors. The original toy didn’t have motors, so he added a new RC gearbox and motors to drive the little fella around. Installed behind Wall-Es eye is a USB webcam. Running behind the scenes is a mumble server that does the audio, a copy of Chromium that shows the video, and an Apache server that feeds the captured video to the other end of the conversation. The whole thing is tied together by a few scripts that kick things off appropriately and allow the user to remotely control Wall-E. It’s a cute build, and hopefully Wall-E can still find his EVE while performing his new corporate duties.