MAIN OFFICE

AIQ-Synertial Ltd

Unit 22 and 23

Sussex Innovation Centre

Science Park Square

Falmer, East Sussex

BN1 9SB

Tel: +44 (020) 3865 3327

AIQ-Synertial, 8F, No.426, Linsen N. Rd., Taipei 10451, Taiwan

+886-2-25210298

© 2020 AiQ Synertial

Get In Touch with HQ

For your bespoke motion capture needs call

+44 (0) 203 865 3327 or email us using the form below.

NO DRIFT MOCAP

The opto-inertial hybrid Mocapsuit eliminates drift

 

by allowing the Vive trackers to take over positioning of the user's root. Then the skeleton must be accurate or there will be constant foot slides.
 
It keeps a track of its whereabouts accurately, repeatedly, meaning absolutely NO DRIFT!

The Vive system can also help with creating accurate skeletons

to eliminate foot slides.

System used here today is a mocapsuit with a pair of 13-sensor gloves..

 
and only a single HTC-Vive Puck tracker on the coccyx/belt, and a second one used as a reference tracker or to calibrate the Vive with, then leaving it on the table and coming back to it after maneuvers, to show the no-drift magic.

 

Note that stabilizing the pelvis with an optical system, gives the data a quality not seen in inertial-only systems until today. Somewhat of an optical data quality to the whole thing. 
The body doesn’t look clunky like the typical inertial suits.

 

Because it's not an inertial suit, It’s a Mocapsuit. It's an opto-inertial Mocapsuit!

The first two takes above were so exciting to watch in real time 
and to record a video of, so much so that we forgot to press the data 
capture button. But they were so smooth and accurate, pretty, 
that we didn't find the heart to chuck them.

 

Here is the video where we remembered to press the capture button. 
I've exported the data in bvh which you can download below 
so you can check it out for yourself.

 

And if you like to stick your own character on it, 
we have also added below an fbx template for that particular 
bvh file (template needs to match the skeleton file in which 
the data was captured, through a T-pose merge). 
I have added another video (snap fbx character onto bvh) 
below on how to load this bvh file onto this fbx and then Merge a character onto it  (as well as how to load additional bvh files).

 

​BVH download link here

FBX download link here

These are the first takes of course, so in the coming days we will pin down 
the final/final v8 skeleton templates and once that is locked somewhat, 
we will make a new cleaner bvh template and make some more takes with the 
new Mocapsuit and will post. In a week or so.

 

Note: that in the video there is a few frame freezes every second 
(it’s not the video app you’re using). That’s our performance enhancing method, 
which doesn’t effect the captured data (visible only in SynDash real-time stream).