Hey @J_Qin, thanks for suggestion! It could work, but my goal here was really to leverage system-provided services instead of rolling my own algorithm. I ended up having a two-step process: room capture, then tap on real world objects.

Hey @J_Qin , thanks for suggestion! It could work, but my goal here was really to leverage system-provided services instead of rolling my own algorithm. I ended up having a two-step process: room capture, then tap on real world objects.

Hi @syncsignal Unfortunately, I could not implement the process of scanning room & marking position of physical objects in a single session. It was okay to implement the feature like this. so we went that route. I did not have the chance to try again lately...

Thanks for your suggestion! It could work, but my goal here was really to leverage system-provided services instead of rolling my own algorithm. I ended up having a two-step process: room capture, then tap on real world objects.

Thanks for your suggestion! It might work, but my goal here was really to leverage system-provided services instead of rolling my own algorithm. I ended up having a two-step process: room capture, then tap on real world objects.

Thanks for your suggestion! It could work, but my goal here was really to leverage system-provided services instead of rolling my own algorithm. I ended up having a two-step process: room capture, then tap on real world objects.