0

I want to develop something like Woorld kind of game with Kudan. So I need to do 2 things:

  1. get the virtual point markers from SLAM frameworks.
  2. Based on the virtual points, build polygons for object occlusions.

I know how to do 2, but I don't know how to get the virtual points from the frameworks.

Is there a way to get the virtual points from the frameworks? Or is there an alternative way to deal with this?

sooon
  • 4,718
  • 8
  • 63
  • 116

1 Answers1

0

Kudan places the camera at origin and based on calculation, the model is placed. So basically you want the object position.

Everts
  • 10,408
  • 2
  • 34
  • 45
  • Do they have a build in frameworks for object occlusion? – sooon Jul 26 '16 at 03:06
  • Cant say for sure but I would say no. From what I experienced, the system relies on hardware for rotation (some android won't work due to lack of gyro). Positioning, I would assume it takes the rotation of the device and an assumed height of user (1.70m), then from there, trig will do the rest. Using accelerometer and gyro and maybe some environment recognition and you got SLAM. Wikitude does a bit of spatial recognition but their 3D dev is beta qnd I doubt it takes care of occlusion. Hololens does proper spatial mapping but is USD3000. – Everts Jul 26 '16 at 20:50