Very interesting that it only requires two light transmitters as opposed to three, since suppose there are two transmitters A and B, an object that they are tracking will be at a distance Da and Db. Given this parameters, there should actually be two possible places in the 3D world space where the object can be, with the two points being symmetrical across the straight path from A to B. This is why I thought a third transmitter is necessary for triangulation. Also, from the diagram, it appears that if the object went beyond the blue box, it would not be able to remain tracked, although mathematically it should still be as long as Da and Db are valid.
muuncakez
This tech reminds me a lot of the Wii and the little tracking bar that would sensor the front end of the wii remote or the xbox kinect that could recognize a hand and used that as a controller instead, super cool to see how far this tech has come now !
dhruvchowdhary
It’s clever how the HTC Vive uses only two light transmitters for tracking, as jayc809 pointed out. But I’m curious, how does the system ensure accuracy and prevent 'drift' over time, especially when players move quickly or block the line of sight to the transmitters?
caelinsutch
Is this drastically more/less accurate than other methods or faster to implement? Curious why Vive went this way for a bit
aidangarde
This solution seems like it would alleviate the tracking issue expressed in previous slides. With both cameras and two static points in the room, tracking headmotion should be pretty precise. I wonder how much the quality would sink if only one transmitter was used. Hypothetically having one still point about the space should be enough to ground the head motion tracker. Vice versa, how much additional accuracy would be introduced by a third transmitter, perhaps on the ground rather than the ceiling.
Very interesting that it only requires two light transmitters as opposed to three, since suppose there are two transmitters A and B, an object that they are tracking will be at a distance Da and Db. Given this parameters, there should actually be two possible places in the 3D world space where the object can be, with the two points being symmetrical across the straight path from A to B. This is why I thought a third transmitter is necessary for triangulation. Also, from the diagram, it appears that if the object went beyond the blue box, it would not be able to remain tracked, although mathematically it should still be as long as Da and Db are valid.
This tech reminds me a lot of the Wii and the little tracking bar that would sensor the front end of the wii remote or the xbox kinect that could recognize a hand and used that as a controller instead, super cool to see how far this tech has come now !
It’s clever how the HTC Vive uses only two light transmitters for tracking, as jayc809 pointed out. But I’m curious, how does the system ensure accuracy and prevent 'drift' over time, especially when players move quickly or block the line of sight to the transmitters?
Is this drastically more/less accurate than other methods or faster to implement? Curious why Vive went this way for a bit
This solution seems like it would alleviate the tracking issue expressed in previous slides. With both cameras and two static points in the room, tracking headmotion should be pretty precise. I wonder how much the quality would sink if only one transmitter was used. Hypothetically having one still point about the space should be enough to ground the head motion tracker. Vice versa, how much additional accuracy would be introduced by a third transmitter, perhaps on the ground rather than the ceiling.