Log in or register to post comments

Show text on point enter I'm unity

May 6, 2016 - 6:04am #1

I have successfully enabled 3d object detection through vuforia in Unity. I have attached a crosshair (reticle) at the centre of the screen in screenspace overlay. when the user moves his phone over the 3d object which is produced upon object detection, I want a label to appear when crosshair crosses different parts of the 3d object. I tried many methods including, collision, cursor and reticle. It is not working. Is there any easy way to implement this so that I can use event trigger pointer enter to make few things happening on the game.

Show text on point enter I'm unity

October 2, 2016 - 9:17am #5

As I have a passion for sniper gaming, I have earned experience in targeting the objects with the help of CS:GO Crosshairs, and I know how to hit the targets.

Show text on point enter I'm unity

May 17, 2016 - 8:07pm #4

It is really seems to be hard and complicated with my littlle knowledge. I have now changed from object detection to planar image detection. Could you please guide me how i can enable pointer in planar image mode with produced 3d image?
thanks

Show text on point enter I'm unity

May 6, 2016 - 9:03pm #3

Thank you so much for a detailed and helpful answer.

I will try the solution you suggested. 

why thought of an easy solution because, upon target detection, the 3d object displayed in Unity has multiple components each with individual mesh collider. Each collider mesh is touch-clickable and I am able to create a touch event and achive the interaction I am trying for. But what I was trying is to use 'gaze' instead of touch. instead of user touching each component and knowing about them, if they point cursor/reticle, I can use pointer enter and exit events to enable/disable labels on each component.

Show text on point enter I'm unity

May 6, 2016 - 3:41pm #2

Hello,

This is challenging as you are using an object target, rather than a planar target. In my opinion, to get this working you'll have to project a 3D point in scene coordinates to screen coordindates for both the object target and the reticle.

For the former, try using the projectPoint() API to transform the scene coordinates to camera pixel coordinates. From there, you can transform the camera coordinates to screen coordinates using settings applied to the Renderer via the VideoBackgroundConfig structure. My expectation is that if you do the same for the reticle, you can then determine where it is pointing on a given object.

What I haven't completely thought through is how to handle areas on your object that may be occluded. Also, I don't believe my solution qualifies for an "easy" implementation.

Good luck.

Log in or register to post comments