oliverkreylos
Latest
Oculus Rift user transplants his whole body into VR using three Kinects
So, not every indie developer abandoned Oculus VR after the Facebook take-over. 3D video expert Oliver Kreylos is one of those who stuck around, and we're mighty glad he did. Check out his three-way-Kinect-plus-Rift setup in the video below and you'll be glad too -- even if the second video, showing the wearer's POV (and intended solely for other Rift owners) makes you retch a couple of times. Video feeds from the three Kinects are merged to create a relatively solid 3D rendition of Kreylos as he moves about inside a VR environment. The merged feed is then sent to the Oculus so that he can look around at everything inside that environment, including his own body. Kreylos says that even though his limbs were a bit fuzzy during the experiment, due to the relatively low resolution of the cameras, the system's low latency allowed his brain to believe that the body parts really belonged to him, to the point where it felt natural to try to reach out and touch non-existent objects. And bear in mind: All of this was achieved using first-gen Kinects, rather than the much more powerful version that is (usually) supplied with the Xbox One, so there's already plenty of scope for improvement.
Two Kinects join forces to make 3D telepresence, enable virtual light saber battles (video)
UC Davis's Oliver Kreylos has been responsible for two of our most impressive Kinect hacks yet. He was one of the first to get proper 3D video out of the thing, following that up by pairing up two of the cameras, one to fill in the gaps of the other. You might have thought he was just playing around but no -- oh no. There was a method to the madness and his ulterior motive has been revealed: 3D telepresence. This is what he's been working on all along and he has an early version operational, using the output from two Kinects in a remote office to beam a 3D representation of another person to his display, which he can navigate around (and through) using a Wiimote. Meanwhile, the viewer can see the position of Oliver in real-time, a virtual camera floating around and enabling them to maintain eye contact despite her not actually looking at either physical camera. That demonstration is embedded after the break along with a somewhat fanciful follow-up in which Kreylos engages in a rather... protracted lightsaber battle against the forces of evil.
Two Kinects join forces to create better 3D video, blow our minds (video)
Remember when Oliver Kreylos impressed and shocked us by showing that a single Microsoft Kinect could create some remarkably stout 3D video? He's back at it again, this time blowing minds and demonstrating that two Kinects can be paired and their output meshed -- one basically filling in the gaps of the other. He found that the two do create some interference, the dotted IR pattern of one causing some holes and blotches in the other, but when the two are combined they basically help each other out and the results are quite impressive. As you can see in the video after the break, Oliver is able to rotate the camera perspective and basically film himself from a new camera angle that exists somewhere in between the position of the two Kinects, and do-so in real-time. Sure, the quality leaves a lot to be desired, but still. Wow. [Thanks, Mohammad]
Hack turns Kinect into 3D video capture tool
We all knew this would inevitably happen, but seeing it in action is something else -- the Kinect transformed by the power of open-source drivers into a true 3D video camera for capturing oneself. UC Davis visualization researcher Oliver Kreylos fed the streams from his peripheral's infrared and color cameras into a custom program that interpolated and reconstructed the result, generating a mildly mindblowing 3D virtual reality environment he can manipulate at will. And if it makes him look a little bit like the proficiently penciled protagonists in Take On Me, that's just the cherry on top. Don't miss the videos after the break to see what we're talking about.
