Items with no label
3335 Discussions

How to track cube position in 3D space using ZR300 ?

BLera
Beginner
1,151 Views

Hi,

I ordered a kit ZR300 + Intel Joule to qualify the ability of the camera to track multiple cubes positions in 3D space.

There are some tutorial recognizing mugs, cups, ... but they don't recognize cubes.

  • Do I have to use the Object Recognition Tool ?
  • Do I have to do myself depth buffer analysis ?
  • Do you have a better alternative to quickly have solid results ?
  • Should I use another camera like R200 or SR300 ?

Thanks in advance.

1 Solution
idata
Employee
263 Views

Hi Valence,

 

 

Thank you for your patience. We now have some information to help you with your request. In regards to the object recognition tool, you can find more information on programming with the YOR Tool at the bottom of this page, https://software.intel.com/en-us/documentation/realsense-for-linux-developers-guide/developing-apps, and here https://software.intel.com/en-us/documentation/realsense-for-linux-developers-guide/yor-tool.

 

 

However, the current functionality won't be able to track a new object in a scene. The released YOR Tool has limited accuracy and also limited to recognition alone without localization (finding the object in the complex scene pointing it with bounding box).

 

 

I hope you find this information useful.

 

 

-Sergio A

 

View solution in original post

0 Kudos
1 Reply
idata
Employee
264 Views

Hi Valence,

 

 

Thank you for your patience. We now have some information to help you with your request. In regards to the object recognition tool, you can find more information on programming with the YOR Tool at the bottom of this page, https://software.intel.com/en-us/documentation/realsense-for-linux-developers-guide/developing-apps, and here https://software.intel.com/en-us/documentation/realsense-for-linux-developers-guide/yor-tool.

 

 

However, the current functionality won't be able to track a new object in a scene. The released YOR Tool has limited accuracy and also limited to recognition alone without localization (finding the object in the complex scene pointing it with bounding box).

 

 

I hope you find this information useful.

 

 

-Sergio A

 

0 Kudos
Reply