by YEI Technology
Developer Kit for the PrioVR - Full-body immersion to interact naturally with virtual worlds.
This project's funding goal was not reached on October 18, 2013.
See full bio
Use this space to cheer the creator along, ask questions, and talk to your fellow backers. Please remember to be respectful and considerate. Thanks!
We're live at --> https://www.kickstarter.com/projects/yeitechnology/priovr-suit-up-game-on
Looking forward to the relaunch.
Saw: http://gizmodo.com/this-awesome-oculus-rift-rig-is-like-real-life-zombie-l-1516510810 and was thinking PrioVR and YEI could do this now, for less, with no external/fixed cameras!
@Hesham and @USM-Valor - We're excited too! Everyone has been working hard this week to get everything together and ready for you!!!!
Looking forward to the reboot. The date is creeping up on us.
I agree with everyone below! Looking forward to the relaunch or just a direct sale :)
@Yang - Not yet, but we're woking on it!
Is there any update?
Please, you have to relaunch D: Every VR controller and project out there keeps ending up about $100 higher than I'd expect to pay, whereas you guys made a decided effort to drop your prices AND are, in my opinion, offering a superior product.
They are coming back !
See their plans etc in the above mentioned interview in Road to VR
Thanks! Glad you are excited! We are too!
We don't have a specific re-launch date yet. We need to regroup and re-evaluate. We will definitely keep you posted as we work out the details!
We are still trying to figure out which languages we will support. Our current 3-Space Sensors SDKs are in C/Python. We will at least have C as a supported language. Hope that helps!
@dyru, @ USM-Valor
We are disappointed too, but this was a great learning experience for us! We are optimistic and looking forward to the re-launch!
Sad to see this was not funded. I would suggest relaunching when there is news of the Oculus Rift being released. That should provide the needed VR user base to ensure this project is successful.
Best of luck!
Not funded? What is wrong with this world!?
6 minutes to go... So! How long should we wait for a revised and relaunched Kickstarter campaign? :D I'll have my finger on the "back this project" button ;)
i noticed you are also working together with Tactical Haptics Reactive Grip Controller for future upgrade. This controller would be a perfect combination with your PrioVR.
Out of curiosity, given that I have not noticed this anywhere else. What programming languages will the SDK support? Will it strictly be C/C++ or will it support other languages?
Hoping to see an amazing community push in the next 11 hours, but if not I'm glad to hear that you will keep at it. Hoping to see another kickstarter personally as opposed to traditional funding.. but whatever works!
@Dennis van der Pool - Definitely something we're working on =)
@Creator: Consider adding a consumer model with the relaunch?
I had to make a choice between PrioVR & STEM system, i choose for the STEM system because they offered a consumer model.
@Trent - The person who can answer that for you is out of the office this week. I'll make sure to get an answer for you when they return!
@Kwesi - Thanks!!!
How flexible are the wires you used? They look like a custom solution, what sort of port is it that is preventing the wires from becoming disconnected or being damaged by continued use? Technical details are welcome!
Awesome product! Keep up the good work YEI Tech.
@Yang Liang @David Blemur @Marco Massarelli @Matt Robinson
Thanks for your support! If this Kickstarter isn’t funded this time around, rest assured that we will do everything we can to make sure you get a system and the PrioVR makes it to market. We are looking into more traditional funding methods, relaunching another campaign, and are looking for more opportunities to get the word out about PrioVR. We haven’t given up yet and don’t plan to anytime soon!
Yes I would buy a prototype too if it was available. It would be good for us developers to get our hands on them to start integrating it into our games.
Ad David says, is there a chance that backers will be allowed to buy a prototype if the campaign is unsuccessful and no new Kickstarter campaign is launched?
Regardless if the goal is reached or not, I still want one.
Good! Even if it does not get funded this time, I won't stop backing up PrioVR until I get mine.
We actually just posted an update about it! We were working on it as you commented! Thanks for you support!
If the Kickstarter does not get funded, we will most likely be relaunching the campaign or funding it in more traditional ways. Even though we only have a few days left, we are still working hard to try and meet our funding goal :)
Also, why is this RoadToVR article not in an update? Just got added today - http://www.roadtovr.com/priovr-prototype-vr-motion-tracking-video-3-space-hands-on :)
What Happens if this doesn't get funded? Continue anyway, repost? Thank you
@Patrick Vidal - Yes, this is correct, we do use the accelerometer and magnetometer to drift-compensate the gyroscope. Our dynamic sensor confidence algorithm, in a nutshell, assigns weights to the accelerometer and magnetometer indicating how trusted each component sensor is at any given time. Whenever the sensor is experiencing motion, it is possible to rely mostly on the gyroscope as it is very accurate during periods of motion, whereas the other two sensors or not. When we detect that the sensor is stationary, we ‘trust’ the accelerometer and magnetometer more. Keep in mind that this does not mean that we ever completely ignore the readings from any component sensor--we only change the influence each sensor has on the overall orientation based on movement.
Our sensors are factory-calibrated and typically do not need to be re-calibrated. In the case that the magnetic environment differs from the original location in which it was calibrated, re-calibration can be performed very easily thanks to a calibration utility that is provided at no charge.
Like any orientation sensor relying on a magnetometer, ours can be adversely influenced by external magnetic fields that are of a different magnitude and direction than the Earth’s ambient field. Unlike other orientation sensors, however, we are able to dynamically detect and mitigate the effects of magnetic perturbations, particularly fields that are induced by objects that move into the sphere of influence of the sensor itself.
@Brian Parker - For most gaming applications, the wiimote could suffice as a single control mechanism and this would be particularly well-suited in conjunction with our PrioVR LZ option. Input nodes are designed to allow for more varied modes of two-way communication, such as providing analog measurements or haptic feedback.
@Peter - Yes, it is possible to move the head without re-orienting any other part of the body and this movement would not be reflected accurately without a sensor placed on the neck. However, in applications where this particular orientation needs to be captured, it is possible to reconfigure the PrioVR system by adding another sensor along the chain. Thanks for the video suggestion as well! We are always brainstorming new and innovative ways to display the power and potential of the PrioVR system and will keep you updated as new demos are developed.
@Steven - Those are all great ideas that I’ll take to the development team. I know they’ve been throwing around the idea of a DDR type demo so I’m glad we’re on the same page!
A few slightly more technical questions...
I assume your 3-space sensor uses the IMU compass/accelerometer readings to counter gyro drift (and the gyro to reduce noise on the compass & accelerometer readings). I am curious about what your "dynamic sensor confidence algorithm" does.
Does the sensor/IMU need to be calibrated?
How sensitive is your system to irregularities in the magnetic field?
What about changes (in the magnetic field) over time?
Would it still work reliably near high power lines or magnets?
I realize these are not typical use conditions. I am mostly asking out of personal curiosity.
@YEI: Oh I'm afraid I not sexy enough there. ;)
Keep the updates coming guys!
@Brian: As far as I gathered the input node would greatly help to keep the entire system mobile and cheaper, as it would enable you to also use cable bound controllers with the WIFI/BT link of the PrioVR .
I guess I dont understand the need for input nodes. The wiimote seems like a perfect controller since it's one handed.
You wouldnt need any kind of input node or tracking of the controller itself since it would be locked to your hand position.
You could use the trigger to grab things and the A button to move in the direction you're facing. I think there's another button that could be used to shoot.
Thank you for taking the time to respond. Wrt, to the precision of the trackers i trust that you know what you are talking about, and the tracking seen in the demo was pretty convincing.
I am however a little confused about the head thing i mentioned. It is true that in order to move my head side-to-side i need to move my neck, but i don't need to re-orient anything else , and as far as i can tell the neck itself does not have a tracker attached?
You almost convinced me to back both this and STEM, but more videos would help a lot. Especially (as already mentioned) a version of the tuscany demo with PrioVR.
Another thing that would be helpful, but is potentially problematic to do, would be to promise some kind of razer hydra / STEM emulation. Anyway, if the kickstarter fails, i hope the campaign will be relaunched.
Hey guys, one word DEMOS.
As a backer for both this and STEM, I have to say this system IS better. Full body is the next step... 5 sensor per user doesn't do much more then the Wii or PS Move. The STEM controller is silly.
I mean, correct me if I'm wrong, but with this system, you can have a ROBO fighting simulation.. elbow, knee or even spin kick without any need for buttons. Forget about buttons and controllers... full body gestures for spells and weapons! Attach one controller to fingers "mitten style" and use that for grabbing...
That demo with the "avoid the box" this was cool but it was not obvious enough how the extra sensors, or maybe it wasn't in a relevant enough setting with today's popular gaming.
As for controllers, I think one can just grab an off the shelf controller and work with it. Show how much more it can do in addition to gaming. How about working out/exercise use. If you can attach the sensors to physical object.. say like a gun or sword.. that would be the killer feature, much more realistic than trying to go with wrist movement only. I would rather hold a bat made of rolled paper with controller in it then pretend some game pad to be a bat.
Do a Miku Hatsune video for fun to get it viral. Do a DDR dance demo! Have it mounted on a rally driver...
I don't know how much cost or work it adds, but being able to add on the fully wireless module would be a plus.
I hope you guys relaunch the campaign if its not reached this time around.
In that particular video, some of the jerkiness results from the usage of raw, unprocessed and unfiltered data, which was intended more to convey the idea rather than demonstrate the results that can be expected from the final system. Additionally, we used some of our early, unrefined pedestrian tracking algorithms, which may contribute some to the overall perceived jitter. Extra lag is also introduced by wireless contention caused by each of the 17 sensors streaming simultaneously. PrioVR will eliminate this problem by using a single wireless communication stream.
P.S. I've cross posted this answer to the update comments as well
@Kevin Cole - Relaunch or direct sale are both ideas being tossed around here (complete with more super awesome video action)!
@Sven H. -
Does that mean you're volunteering to be in the next video? =P
Our plan is to directly support OEM and 3rd party controllers like the Wii Nunchuck and Wii classic game controllers. By supporting these commercially available controllers, we can keep costs down and allow a range of price/performance/availability options for users to select from. Also, the hub unit for all PrioVR system versions are planned to have a port to allow connection of these controllers, so no additional input nodes would be needed for controller support. We have these controllers working with PrioVR sensors right now in the lab and hope to have a video posted shortly.
1. While you can move your head side-to-side without changing its orientation, it is impossible to move the head side to side without changing the orientation of the neck, torso, legs, feet, or some other part of the body. Thus, a body fully outfitted with orientation trackers and a skeletal constrain model can fully resolve the complete pose of the player accurately.
2. You are correct that the accuracy (or at least repeatability) of the sensors is important to performance as is the physical mounting method. The sensors used in the PrioVR exhibit performance characteristics that are comparable to or better than competing inertial AHRS systems costing in the $1500 - $2000 per sensor range. The accuracy is +/- 1 degree average and the repeatability is 0.085 degrees. Given these performance characteristics, combined with our new sensor attachment design, the sensors will do quite well at having repeated motions and poses produce identical results time after time. Thus, if your hands are together and you wave your arms around for an hour or two, then your hands will be together after that when placed in an identical position.
Much like @Kevin Cole, I'd buy the head-to-head if relaunch or sell separately in a heartbeat! Also, I just wanted to ask about the demo, it was a bit jumpy and discontinuous in parts for the tracking, is that b/c this is a prototype or prototype software or an artifact of the recording? Still want this and the glove which you teased might be mentioned later, please do!
I hope you relaunch this campain. Or perhaps offer the priovr system for sale separately. Even when the initial backer cost was $600 that was still $400 less than the full motion capture option you had with
Your other sensors you currently sell today.
I backed both projects because i saw them as a set of compatible tech that would have been awesome to combine.
As it stands, the priovr would be the best for the omni treadmill.
Anyway, you're shy by $140k currently with 9 days remaining. I really hope you launch again with better promotional videos illustrating the true wins of your tech.
I recommend videos placing you in the omni-treadmill, vids displaying full motion capture, combining the system with stem to work as a team even for even more accurate tracking to take advantage of stems positional tracking and fill in the gaps of its IK limitations. Even more vids for the indie game devs as far as better mocap.
Right now its like watching betamax vs vhs. Although your product may be higher quality, the direct involvement of oculus and all the media hype of stem is really what is making them win.